isaac-sim / omniisaacgymenvs

Reinforcement Learning Environments for Omniverse Isaac Gym

Omniverse Isaac Gym Reinforcement Learning Environments for Isaac Sim

PLEASE NOTE: Version 4.0.0 will be the last release of OmniIsaacGymEnvs. Moving forward, OmniIsaacGymEnvs will be merging with IsaacLab (https://github.com/isaac-sim/IsaacLab). All future updates will be available as part of the IsaacLab repository.

For tutorials on migrating to IsaacLab, please visit: https://isaac-sim.github.io/IsaacLab/source/migration/migrating_from_omniisaacgymenvs.html.

About this repository

This repository contains Reinforcement Learning examples that can be run with the latest release of Isaac Sim. RL examples are trained using PPO from rl_games library and examples are built on top of Isaac Sim's omni.isaac.core and omni.isaac.gym frameworks.

Please see release notes for the latest updates.

 

 

 

 

 

System Requirements

It is recommended to have at least 32GB RAM and a GPU with at least 12GB VRAM. For detailed system requirements, please visit the Isaac Sim System Requirements page. Please refer to the Troubleshooting page for a detailed breakdown of memory consumption.

Installation

Follow the Isaac Sim documentation to install the latest Isaac Sim release.

Examples in this repository rely on features from the most recent Isaac Sim release. Please make sure to update any existing Isaac Sim build to the latest release version, 4.0.0, to ensure examples work as expected.

Once installed, this repository can be used as a python module, omniisaacgymenvs, with the python executable provided in Isaac Sim.

To install omniisaacgymenvs, first clone this repository:

git clone https://github.com/NVIDIA-Omniverse/OmniIsaacGymEnvs.git

Once cloned, locate the python executable in Isaac Sim. By default, this should be python.sh. We will refer to this path as PYTHON_PATH.

To set a PYTHON_PATH variable in the terminal that links to the python executable, we can run a command that resembles the following. Make sure to update the paths to your local path.

For Linux: alias PYTHON_PATH=~/.local/share/ov/pkg/isaac_sim-*/python.sh
For Windows: doskey PYTHON_PATH=C:\Users\user\AppData\Local\ov\pkg\isaac_sim-*\python.bat $*
For IsaacSim Docker: alias PYTHON_PATH=/isaac-sim/python.sh

Install omniisaacgymenvs as a python module for PYTHON_PATH:

PYTHON_PATH -m pip install -e .

The following error may appear during the initial installation. This error is harmless and can be ignored.

ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.

Running the examples

Note: All commands should be executed from OmniIsaacGymEnvs/omniisaacgymenvs.

To train your first policy, run:

PYTHON_PATH scripts/rlgames_train.py task=Cartpole

An Isaac Sim app window should be launched. Once Isaac Sim initialization completes, the Cartpole scene will be constructed and simulation will start running automatically. The process will terminate once training finishes.

Note that by default, we show a Viewport window with rendering, which slows down training. You can choose to close the Viewport window during training for better performance. The Viewport window can be re-enabled by selecting Window > Viewport from the top menu bar.

To achieve maximum performance, launch training in headless mode as follows:

PYTHON_PATH scripts/rlgames_train.py task=Ant headless=True

A Note on the Startup Time of the Simulation

Some of the examples could take a few minutes to load because the startup time scales based on the number of environments. The startup time will continually be optimized in future releases.

Extension Workflow

The extension workflow provides a simple user interface for creating and launching RL tasks. To launch Isaac Sim for the extension workflow, run:

./<isaac_sim_root>/isaac-sim.gym.sh --ext-folder </parent/directory/to/OIGE>

Note: isaac_sim_root should be located in the same directory as python.sh.

The UI window can be activated from Isaac Examples > RL Examples by navigating the top menu bar. For more details on the extension workflow, please refer to the documentation.

If you are running into an issue where the Isaac Examples menu is missing, please modify OmniIsaacGymEnvs/omniisaacgymenvs/__init__.py as follow to expose the underlying error:

import traceback
try:
    from .extension import RLExtension, get_instance
except Exception:
    print(traceback.format_exc())

In the case of a ModuleNotFoundError for hydra, please check your C:\Users\user\AppData\Roaming\Python\Python310 directory and remove any site-packages directory that may contain the hydra package. Then, re-run the pip install -e . command for OmniIsaacGymEnvs.

Loading trained models // Checkpoints

Checkpoints are saved in the folder runs/EXPERIMENT_NAME/nn where EXPERIMENT_NAME defaults to the task name, but can also be overridden via the experiment argument.

To load a trained checkpoint and continue training, use the checkpoint argument:

PYTHON_PATH scripts/rlgames_train.py task=Ant checkpoint=runs/Ant/nn/Ant.pth

To load a trained checkpoint and only perform inference (no training), pass test=True as an argument, along with the checkpoint name. To avoid rendering overhead, you may also want to run with fewer environments using num_envs=64:

PYTHON_PATH scripts/rlgames_train.py task=Ant checkpoint=runs/Ant/nn/Ant.pth test=True num_envs=64

Note that if there are special characters such as [ or = in the checkpoint names, you will need to escape them and put quotes around the string. For example, checkpoint="runs/Ant/nn/last_Antep\=501rew\[5981.31\].pth"

We provide pre-trained checkpoints on the Nucleus server under Assets/Isaac/4.0/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints. Run the following command to launch inference with pre-trained checkpoint:

Localhost (To set up localhost, please refer to the Isaac Sim installation guide):

PYTHON_PATH scripts/rlgames_train.py task=Ant checkpoint=omniverse://localhost/NVIDIA/Assets/Isaac/4.0/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints/ant.pth test=True num_envs=64

Production server:

PYTHON_PATH scripts/rlgames_train.py task=Ant checkpoint=http://omniverse-content-production.s3-us-west-2.amazonaws.com/Assets/Isaac/4.0/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints/ant.pth test=True num_envs=64

When running with a pre-trained checkpoint for the first time, we will automatically download the checkpoint file to omniisaacgymenvs/checkpoints. For subsequent runs, we will re-use the file that has already been downloaded, and will not overwrite existing checkpoints with the same name in the checkpoints folder.

Runing from Docker

Latest Isaac Sim Docker image can be found on NGC. A utility script is provided at docker/run_docker.sh to help initialize this repository and launch the Isaac Sim docker container. The script can be run with:

./docker/run_docker.sh

Then, training can be launched from the container with:

/isaac-sim/python.sh scripts/rlgames_train.py headless=True task=Ant

To run the Isaac Sim docker with UI, use the following script:

./docker/run_docker_viewer.sh

Then, training can be launched from the container with:

/isaac-sim/python.sh scripts/rlgames_train.py task=Ant

To avoid re-installing OIGE each time a container is launched, we also provide a dockerfile that can be used to build an image with OIGE installed. To build the image, run:

docker build -t isaac-sim-oige -f docker/dockerfile .

Then, start a container with the built image:

./docker/run_dockerfile.sh

Then, training can be launched from the container with:

/isaac-sim/python.sh scripts/rlgames_train.py task=Ant headless=True

Isaac Sim Automator

Cloud instances for AWS, Azure, or GCP can be setup using Isaac Automator.

Livestream

OmniIsaacGymEnvs supports livestream through the Omniverse Streaming Client. To enable this feature, add the commandline argument enable_livestream=True:

PYTHON_PATH scripts/rlgames_train.py task=Ant headless=True enable_livestream=True

Connect from the Omniverse Streaming Client once the SimulationApp has been created. Note that enabling livestream is equivalent to training with the viewer enabled, thus the speed of training/inferencing will decrease compared to running in headless mode.

Training Scripts

All scripts provided in omniisaacgymenvs/scripts can be launched directly with PYTHON_PATH.

To test out a task without RL in the loop, run the random policy script with:

PYTHON_PATH scripts/random_policy.py task=Cartpole

This script will sample random actions from the action space and apply these actions to your task without running any RL policies. Simulation should start automatically after launching the script, and will run indefinitely until terminated.

To run a simple form of PPO from rl_games, use the single-threaded training script:

PYTHON_PATH scripts/rlgames_train.py task=Cartpole

This script creates an instance of the PPO runner in rl_games and automatically launches training and simulation. Once training completes (the total number of iterations have been reached), the script will exit. If running inference with test=True checkpoint=<path/to/checkpoint>, the script will run indefinitely until terminated. Note that this script will have limitations on interaction with the UI.

Configuration and command line arguments

We use Hydra to manage the config.

Common arguments for the training scripts are:

• task=TASK - Selects which task to use. Any of AllegroHand, Ant, Anymal, AnymalTerrain, BallBalance, Cartpole, CartpoleCamera, Crazyflie, FactoryTaskNutBoltPick, FactoryTaskNutBoltPlace, FactoryTaskNutBoltScrew, FrankaCabinet, FrankaDeformable, Humanoid, Ingenuity, Quadcopter, ShadowHand, ShadowHandOpenAI_FF, ShadowHandOpenAI_LSTM (these correspond to the config for each environment in the folder omniisaacgymenvs/cfg/task)
• train=TRAIN - Selects which training config to use. Will automatically default to the correct config for the environment (ie. <TASK>PPO).
• num_envs=NUM_ENVS - Selects the number of environments to use (overriding the default number of environments set in the task config).
• seed=SEED - Sets a seed value for randomization, and overrides the default seed in the task config
• pipeline=PIPELINE - Which API pipeline to use. Defaults to gpu, can also set to cpu. When using the gpu pipeline, all data stays on the GPU. When using the cpu pipeline, simulation can run on either CPU or GPU, depending on the sim_device setting, but a copy of the data is always made on the CPU at every step.
• sim_device=SIM_DEVICE - Device used for physics simulation. Set to gpu (default) to use GPU and to cpu for CPU.
• device_id=DEVICE_ID - Device ID for GPU to use for simulation and task. Defaults to 0. This parameter will only be used if simulation runs on GPU.
• rl_device=RL_DEVICE - Which device / ID to use for the RL algorithm. Defaults to cuda:0, and follows PyTorch-like device syntax.
• multi_gpu=MULTI_GPU - Whether to train using multiple GPUs. Defaults to False. Note that this option is only available with rlgames_train.py.
• test=TEST- If set to True, only runs inference on the policy and does not do any training.
• checkpoint=CHECKPOINT_PATH - Path to the checkpoint to load for training or testing.
• headless=HEADLESS - Whether to run in headless mode.
• enable_livestream=ENABLE_LIVESTREAM - Whether to enable Omniverse streaming.
• experiment=EXPERIMENT - Sets the name of the experiment.
• max_iterations=MAX_ITERATIONS - Sets how many iterations to run for. Reasonable defaults are provided for the provided environments.
• warp=WARP - If set to True, launch the task implemented with Warp backend (Note: not all tasks have a Warp implementation).
• kit_app=KIT_APP - Specifies the absolute path to the kit app file to be used.
• enable_recording=ENABLE_RECORDING - Enables viewport recording while running a task
• recording_interval=RECORDING_INTERVAL - Number of RL steps in between recording sequences. By default, recordings happen every 2000 RL steps.
• recording_length=RECORDING_LENGTH - Number of RL steps to record for each clip. By default, recordings are 100 steps in length.
• recording_fps=RECORDING_FPS - Output FPS of the recorded video. Defaults to 30.
• recording_dir=RECORDING_DIR - Directory to save recordings in. Defaults to the experiment directory.

Hydra also allows setting variables inside config files directly as command line arguments. As an example, to set the minibatch size for a rl_games training run, you can use train.params.config.minibatch_size=64. Similarly, variables in task configs can also be set. For example, task.env.episodeLength=100.

Hydra Notes

Default values for each of these are found in the omniisaacgymenvs/cfg/config.yaml file.

The way that the task and train portions of the config works are through the use of config groups. You can learn more about how these work here The actual configs for task are in omniisaacgymenvs/cfg/task/<TASK>.yaml and for train in omniisaacgymenvs/cfg/train/<TASK>PPO.yaml.

In some places in the config you will find other variables referenced (for example, num_actors: ${....task.env.numEnvs}). Each . represents going one level up in the config hierarchy. This is documented fully here.

Tensorboard

Tensorboard can be launched during training via the following command:

PYTHON_PATH -m tensorboard.main --logdir runs/EXPERIMENT_NAME/summaries

WandB support

You can run (WandB)[https://wandb.ai/] with OmniIsaacGymEnvs by setting wandb_activate=True flag from the command line. You can set the group, name, entity, and project for the run by setting the wandb_group, wandb_name, wandb_entity and wandb_project arguments. Make sure you have WandB installed in the Isaac Sim Python executable with PYTHON_PATH -m pip install wandb before activating.

Training with Multiple GPUs

To train with multiple GPUs, use the following command, where --proc_per_node represents the number of available GPUs:

PYTHON_PATH -m torch.distributed.run --nnodes=1 --nproc_per_node=2 scripts/rlgames_train.py headless=True task=Ant multi_gpu=True

Multi-Node Training

To train across multiple nodes/machines, it is required to launch an individual process on each node. For the master node, use the following command, where --proc_per_node represents the number of available GPUs, and --nnodes represents the number of nodes:

PYTHON_PATH -m torch.distributed.run --nproc_per_node=2 --nnodes=2 --node_rank=0 --rdzv_id=123 --rdzv_backend=c10d --rdzv_endpoint=localhost:5555 scripts/rlgames_train.py headless=True task=Ant multi_gpu=True

Note that the port (5555) can be replaced with any other available port.

For non-master nodes, use the following command, replacing --node_rank with the index of each machine:

PYTHON_PATH -m torch.distributed.run --nproc_per_node=2 --nnodes=2 --node_rank=1 --rdzv_id=123 --rdzv_backend=c10d --rdzv_endpoint=ip_of_master_machine:5555 scripts/rlgames_train.py headless=True task=Ant multi_gpu=True

For more details on multi-node training with PyTorch, please visit here. As mentioned in the PyTorch documentation, "multinode training is bottlenecked by inter-node communication latencies". When this latency is high, it is possible multi-node training will perform worse than running on a single node instance.

Tasks

Source code for tasks can be found in omniisaacgymenvs/tasks.

Each task follows the frameworks provided in omni.isaac.core and omni.isaac.gym in Isaac Sim.

Refer to docs/framework/framework.md for how to create your own tasks.

Full details on each of the tasks available can be found in the RL examples documentation.

Demo

We provide an interactable demo based on the AnymalTerrain RL example. In this demo, you can click on any of the ANYmals in the scene to go into third-person mode and manually control the robot with your keyboard as follows:

• Up Arrow: Forward linear velocity command
• Down Arrow: Backward linear velocity command
• Left Arrow: Leftward linear velocity command
• Right Arrow: Rightward linear velocity command
• Z: Counterclockwise yaw angular velocity command
• X: Clockwise yaw angular velocity command
• C: Toggles camera view between third-person and scene view while maintaining manual control
• ESC: Unselect a selected ANYmal and yields manual control

Launch this demo with the following command. Note that this demo limits the maximum number of ANYmals in the scene to 128.

PYTHON_PATH scripts/rlgames_demo.py task=AnymalTerrain num_envs=64 checkpoint=omniverse://localhost/NVIDIA/Assets/Isaac/4.0/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints/anymal_terrain.pth

omniisaacgymenvs's Issues

Getting error "No matching distribution found for ray<2.0.0,>=1.11.0" on Windows 11

I've tried installing this on three Windows machines and same error each time

DIfference against IsaacGymEnvs

Hi,

What is the difference between this repo and IsaacGymEnvs?

Multi Gpu training of AnimalTerrain crashes

// using python here because I am using conda

python -m torch.distributed.run --nnodes=1 --nproc_per_node=2 scripts/rlgames_train.py headless=True task=AnymalTerrain multi_gpu=True

WARNING:main:

---

Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.

---

/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config': Defaults list is missing _self_. See https://hydra.cc/docs/upgrades/1.0_to_1.1/default_composition_order for more information
warnings.warn(msg, UserWarning)
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:412: UserWarning: In config: Invalid overriding of hydra/job_logging:
Default list overrides requires 'override' keyword.
See https://hydra.cc/docs/next/upgrades/1.0_to_1.1/defaults_list_override for more information.

deprecation_warning(msg)
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config': Defaults list is missing _self_. See https://hydra.cc/docs/upgrades/1.0_to_1.1/default_composition_order for more information
warnings.warn(msg, UserWarning)
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:412: UserWarning: In config: Invalid overriding of hydra/job_logging:
Default list overrides requires 'override' keyword.
See https://hydra.cc/docs/next/upgrades/1.0_to_1.1/defaults_list_override for more information.

deprecation_warning(msg)
[Warning] [omni.isaac.kit.simulation_app] Modules: ['omni.isaac.kit.app_framework'] were loaded before SimulationApp was started and might not be loaded correctly.
[Warning] [omni.isaac.kit.simulation_app] Please check to make sure no extra omniverse or pxr modules are imported before the call to SimulationApp(...)
Starting kit application with the following args: ['/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.isaac.kit/omni/isaac/kit/simulation_app.py', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/apps/omni.isaac.sim.python.gym.headless.kit', '--/app/tokens/exe-path=/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/kit', '--/persistent/app/viewport/displayOptions=3094', '--/rtx/materialDb/syncLoads=True', '--/rtx/hydra/materialSyncLoads=True--/omni.kit.plugin/syncUsdLoads=True', '--/app/renderer/resolution/width=1280', '--/app/renderer/resolution/height=720', '--/app/window/width=1440', '--/app/window/height=900', '--/renderer/multiGpu/enabled=True', '--/app/fastShutdown=True', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/apps', '--/physics/cudaDevice=1', '--portable', '--no-window']
Passing the following args to the base kit application: ['headless=True', 'task=AnymalTerrain', 'multi_gpu=True']
[Warning] [omni.kit.app.plugin] No crash reporter present, dumps uploading isn't available.
[Info] [carb] Logging to file: /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/logs/Kit/Isaac-Sim/2022.2/kit_20230512_104301.log
2023-05-12 20:43:01 [5ms] [Warning] [omni.ext.plugin] [ext: omni.sensors.nv.lidar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.lidar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.lidar/config'
2023-05-12 20:43:01 [6ms] [Warning] [omni.ext.plugin] [ext: omni.sensors.nv.radar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.radar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.radar/config'
[Warning] [omni.isaac.kit.simulation_app] Modules: ['omni.isaac.kit.app_framework'] were loaded before SimulationApp was started and might not be loaded correctly.
[Warning] [omni.isaac.kit.simulation_app] Please check to make sure no extra omniverse or pxr modules are imported before the call to SimulationApp(...)
Starting kit application with the following args: ['/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.isaac.kit/omni/isaac/kit/simulation_app.py', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/apps/omni.isaac.sim.python.gym.headless.kit', '--/app/tokens/exe-path=/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/kit', '--/persistent/app/viewport/displayOptions=3094', '--/rtx/materialDb/syncLoads=True', '--/rtx/hydra/materialSyncLoads=True--/omni.kit.plugin/syncUsdLoads=True', '--/app/renderer/resolution/width=1280', '--/app/renderer/resolution/height=720', '--/app/window/width=1440', '--/app/window/height=900', '--/renderer/multiGpu/enabled=True', '--/app/fastShutdown=True', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/apps', '--/physics/cudaDevice=0', '--portable', '--no-window']
Passing the following args to the base kit application: ['headless=True', 'task=AnymalTerrain', 'multi_gpu=True']
[Warning] [omni.kit.app.plugin] No crash reporter present, dumps uploading isn't available.
[Info] [carb] Logging to file: /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/logs/Kit/Isaac-Sim/2022.2/kit_20230512_104301.log
2023-05-12 20:43:01 [8ms] [Warning] [omni.ext.plugin] [ext: omni.sensors.nv.lidar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.lidar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.lidar/config'
2023-05-12 20:43:01 [8ms] [Warning] [omni.ext.plugin] [ext: omni.sensors.nv.radar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.radar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.sensors.nv.radar/config'
[0.098s] [ext: omni.stats-0.0.0] startup
[0.103s] [ext: omni.rtx.shadercache-1.0.0] startup
[0.105s] [ext: omni.assets.plugins-0.0.0] startup
[0.106s] [ext: omni.gpu_foundation-0.0.0] startup
[0.111s] [ext: carb.windowing.plugins-1.0.0] startup
2023-05-12 20:43:01 [111ms] [Warning] [carb.windowing-glfw.gamepad] Joystick with unknown remapping detected (will be ignored): ASRock LED Controller [03000000ce260000a201000010010000]
[0.118s] [ext: omni.kit.renderer.init-0.0.0] startup
[0.101s] [ext: omni.stats-0.0.0] startup
[0.105s] [ext: omni.rtx.shadercache-1.0.0] startup
[0.107s] [ext: omni.assets.plugins-0.0.0] startup
[0.108s] [ext: omni.gpu_foundation-0.0.0] startup
[0.113s] [ext: carb.windowing.plugins-1.0.0] startup
2023-05-12 20:43:01 [105ms] [Warning] [carb.windowing-glfw.gamepad] Joystick with unknown remapping detected (will be ignored): ASRock LED Controller [03000000ce260000a201000010010000]
[0.119s] [ext: omni.kit.renderer.init-0.0.0] startup

|---------------------------------------------------------------------------------------------|
| Driver Version: 530.30.02 | Graphics API: Vulkan
|=============================================================================================|
| GPU | Name | Active | LDA | GPU Memory | Vendor-ID | LUID |
| | | | | | Device-ID | UUID |
|---------------------------------------------------------------------------------------------|
| 0 | NVIDIA GeForce RTX 4090 | Yes: 1 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 527cadd8.. |
|---------------------------------------------------------------------------------------------|
| 1 | NVIDIA GeForce RTX 4090 | Yes: 0 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 7c9328c9.. |
|=============================================================================================|
| OS: Linux dl, Version: 5.15.0-69-generic
| XServer Vendor: The X.Org Foundation, XServer Version: 12013000 (1.20.13.0)
| Processor: AMD Ryzen Threadripper PRO 5975WX 32-Cores | Cores: Unknown | Logical: 64
|---------------------------------------------------------------------------------------------|
| Total Memory (MB): 257558 | Free Memory: 249198
| Total Page/Swap (MB): 2047 | Free Page/Swap: 2047
|---------------------------------------------------------------------------------------------|

|---------------------------------------------------------------------------------------------|
| Driver Version: 530.30.02 | Graphics API: Vulkan
|=============================================================================================|
| GPU | Name | Active | LDA | GPU Memory | Vendor-ID | LUID |
| | | | | | Device-ID | UUID |
|---------------------------------------------------------------------------------------------|
| 0 | NVIDIA GeForce RTX 4090 | Yes: 1 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 527cadd8.. |
|---------------------------------------------------------------------------------------------|
| 1 | NVIDIA GeForce RTX 4090 | Yes: 0 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 7c9328c9.. |
|=============================================================================================|
| OS: Linux dl, Version: 5.15.0-69-generic
| XServer Vendor: The X.Org Foundation, XServer Version: 12013000 (1.20.13.0)
| Processor: AMD Ryzen Threadripper PRO 5975WX 32-Cores | Cores: Unknown | Logical: 64
|---------------------------------------------------------------------------------------------|
| Total Memory (MB): 257558 | Free Memory: 249221
| Total Page/Swap (MB): 2047 | Free Page/Swap: 2047
|---------------------------------------------------------------------------------------------|
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] IOMMU is enabled. Found 50 items in /sys/kernel/iommu_groups/.
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] On Linux only, CUDA and the display driver does not support IOMMU-enabled bare-metal PCIe peer to peer memory copy.
However, CUDA and the display driver does support IOMMU via VM pass through. As a consequence, users on Linux,
when running on a native bare metal system, should disable the IOMMU. The IOMMU should be enabled and the VFIO driver
be used as a PCIe pass through for virtual machines.
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin]
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] !!!!! Local system validation failed! Incorrect configuration detected.
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] Summary below. Details above.
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin]
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] - ECC: OK
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] - IOMMU: FAILED
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin]
2023-05-12 20:43:02 [885ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] IOMMU is enabled. Found 50 items in /sys/kernel/iommu_groups/.
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] On Linux only, CUDA and the display driver does not support IOMMU-enabled bare-metal PCIe peer to peer memory copy.
However, CUDA and the display driver does support IOMMU via VM pass through. As a consequence, users on Linux,
when running on a native bare metal system, should disable the IOMMU. The IOMMU should be enabled and the VFIO driver
be used as a PCIe pass through for virtual machines.
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin]
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] !!!!! Local system validation failed! Incorrect configuration detected.
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] Summary below. Details above.
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin]
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] - ECC: OK
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] - IOMMU: FAILED
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin]
2023-05-12 20:43:02 [924ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
[1.758s] [ext: omni.kit.pipapi-0.0.0] startup
[1.761s] [ext: omni.kit.pip_archive-0.0.0] startup
[1.762s] [ext: omni.kit.loop-isaac-1.0.0] startup
[1.780s] [ext: omni.kit.pipapi-0.0.0] startup
[1.763s] [ext: omni.kit.async_engine-0.0.0] startup
[1.763s] [ext: omni.kit.test-0.0.0] startup
[1.784s] [ext: omni.kit.pip_archive-0.0.0] startup
[1.786s] [ext: omni.kit.loop-isaac-1.0.0] startup
[1.786s] [ext: omni.kit.async_engine-0.0.0] startup
[1.787s] [ext: omni.kit.test-0.0.0] startup
[1.775s] [ext: omni.usd.config-1.0.0] startup
[1.778s] [ext: omni.usd.libs-1.0.0] startup
[1.802s] [ext: omni.usd.config-1.0.0] startup
[1.807s] [ext: omni.usd.libs-1.0.0] startup
[1.943s] [ext: omni.isaac.core_archive-2.0.1] startup
[1.948s] [ext: omni.pip.torch-1_13_1-0.1.4] startup
[1.949s] [ext: omni.isaac.ml_archive-1.1.0] startup
[1.949s] [ext: omni.client-0.1.1] startup
[1.956s] [ext: omni.appwindow-1.0.1] startup
[1.957s] [ext: omni.kit.renderer.core-0.0.0] startup
[1.960s] [ext: omni.kit.renderer.capture-0.0.0] startup
[1.961s] [ext: omni.kit.renderer.imgui-0.0.0] startup
[1.981s] [ext: omni.isaac.core_archive-2.0.1] startup
[1.986s] [ext: omni.pip.torch-1_13_1-0.1.4] startup
[1.987s] [ext: omni.isaac.ml_archive-1.1.0] startup
[1.987s] [ext: omni.client-0.1.1] startup
[1.996s] [ext: omni.appwindow-1.0.1] startup
[1.997s] [ext: omni.kit.renderer.core-0.0.0] startup
[2.000s] [ext: omni.kit.renderer.capture-0.0.0] startup
[2.002s] [ext: omni.kit.renderer.imgui-0.0.0] startup
[2.011s] [ext: carb.audio-0.1.0] startup
[2.050s] [ext: carb.audio-0.1.0] startup
[2.037s] [ext: omni.ui-2.14.4] startup
[2.045s] [ext: omni.uiaudio-1.0.0] startup
[2.046s] [ext: omni.kit.mainwindow-1.0.0] startup
[2.047s] [ext: omni.kit.uiapp-0.0.0] startup
[2.047s] [ext: omni.usd.schema.physics-1.0.0] startup
[2.071s] [ext: omni.ui-2.14.4] startup
[2.080s] [ext: omni.uiaudio-1.0.0] startup
[2.081s] [ext: omni.kit.mainwindow-1.0.0] startup
[2.082s] [ext: omni.kit.uiapp-0.0.0] startup
[2.082s] [ext: omni.usd.schema.physics-1.0.0] startup
[2.081s] [ext: omni.usd.schema.geospatial-0.0.0] startup
[2.085s] [ext: omni.usd.schema.audio-0.0.0] startup
[2.089s] [ext: omni.usd.schema.anim-0.0.0] startup
[2.118s] [ext: omni.usd.schema.geospatial-0.0.0] startup
[2.124s] [ext: omni.usd.schema.audio-0.0.0] startup
[2.127s] [ext: omni.usd.schema.anim-0.0.0] startup
[2.129s] [ext: omni.usd.schema.semantics-0.0.0] startup
[2.132s] [ext: omni.usd.schema.physx-0.0.0] startup
[2.169s] [ext: omni.usd.schema.semantics-0.0.0] startup
[2.152s] [ext: omni.usd.schema.omnigraph-1.0.0] startup
[2.172s] [ext: omni.usd.schema.physx-0.0.0] startup
[2.158s] [ext: omni.usd.schema.omniscripting-1.0.0] startup
[2.163s] [ext: omni.kit.window.popup_dialog-2.0.16] startup
[2.166s] [ext: omni.kit.actions.core-1.0.0] startup
[2.167s] [ext: omni.kit.widget.nucleus_connector-1.0.3] startup
[2.169s] [ext: omni.kit.commands-1.4.5] startup
[2.171s] [ext: omni.gpucompute.plugins-0.0.0] startup
[2.172s] [ext: omni.usd.core-1.0.4] startup
[2.175s] [ext: omni.timeline-1.0.5] startup
[2.176s] [ext: omni.hydra.scene_delegate-0.3.0] startup
[2.195s] [ext: omni.usd.schema.omnigraph-1.0.0] startup
[2.182s] [ext: omni.kit.audiodeviceenum-1.0.0] startup
[2.183s] [ext: omni.hydra.usdrt_delegate-4.3.2] startup
[2.201s] [ext: omni.usd.schema.omniscripting-1.0.0] startup
[2.206s] [ext: omni.kit.window.popup_dialog-2.0.16] startup
[2.209s] [ext: omni.kit.actions.core-1.0.0] startup
[2.211s] [ext: omni.kit.widget.nucleus_connector-1.0.3] startup
[2.213s] [ext: omni.kit.commands-1.4.5] startup
[2.216s] [ext: omni.gpucompute.plugins-0.0.0] startup
[2.216s] [ext: omni.usd.core-1.0.4] startup
[2.200s] [ext: omni.graph.tools-1.17.2] startup
[2.220s] [ext: omni.timeline-1.0.5] startup
[2.221s] [ext: omni.hydra.scene_delegate-0.3.0] startup
[2.227s] [ext: omni.kit.audiodeviceenum-1.0.0] startup
[2.228s] [ext: omni.hydra.usdrt_delegate-4.3.2] startup
[2.211s] [ext: omni.usd-1.6.30] startup
[2.245s] [ext: omni.graph.tools-1.17.2] startup
[2.257s] [ext: omni.usd-1.6.30] startup
[2.251s] [ext: omni.kit.collaboration.channel_manager-1.0.9] startup
[2.252s] [ext: omni.kvdb-0.0.0] startup
[2.253s] [ext: omni.kit.usd.layers-2.0.11] startup
[2.260s] [ext: omni.kit.menu.utils-1.4.7] startup
[2.267s] [ext: omni.localcache-0.0.0] startup
[2.268s] [ext: omni.kit.primitive.mesh-1.0.8] startup
[2.270s] [ext: omni.convexdecomposition-104.2.4-5.1] startup
[2.271s] [ext: omni.kit.stage_templates-1.1.13] startup
[2.273s] [ext: omni.kit.usd_undo-0.1.2] startup
[2.273s] [ext: omni.graph.core-2.65.4] startup
[2.276s] [ext: omni.usdphysics-104.2.4-5.1] startup
[2.279s] [ext: omni.graph-1.50.2] startup
[2.300s] [ext: omni.kit.collaboration.channel_manager-1.0.9] startup
[2.301s] [ext: omni.kvdb-0.0.0] startup
[2.303s] [ext: omni.kit.usd.layers-2.0.11] startup
[2.310s] [ext: omni.kit.menu.utils-1.4.7] startup
[2.316s] [ext: omni.localcache-0.0.0] startup
[2.317s] [ext: omni.kit.primitive.mesh-1.0.8] startup
[2.320s] [ext: omni.convexdecomposition-104.2.4-5.1] startup
[2.321s] [ext: omni.kit.stage_templates-1.1.13] startup
[2.323s] [ext: omni.kit.usd_undo-0.1.2] startup
[2.323s] [ext: omni.graph.core-2.65.4] startup
[2.326s] [ext: omni.usdphysics-104.2.4-5.1] startup
[2.329s] [ext: omni.graph-1.50.2] startup
[2.371s] [ext: omni.physx-104.2.4-5.1] startup
[2.404s] [ext: omni.kit.numpy.common-0.1.0] startup
[2.422s] [ext: omni.physx-104.2.4-5.1] startup
[2.405s] [ext: omni.graph.nodes-1.48.3] startup
[2.417s] [ext: omni.isaac.dynamic_control-1.2.3] startup
2023-05-12 20:43:04 [2,431ms] [Warning] [omni.kvdb.plugin] wasn't able to load the meta database, trying to repair it ...
[2.426s] [ext: omni.isaac.kit-1.4.1] startup
[2.426s] [ext: omni.kit.widget.path_field-2.0.4] startup
[2.427s] [ext: omni.kit.search_core-1.0.2] startup
[2.427s] [ext: omni.kit.widget.browser_bar-2.0.5] startup
[2.428s] [ext: omni.kit.widget.filebrowser-2.3.10] startup
[2.436s] [ext: omni.kit.widget.versioning-1.3.8] startup
[2.437s] [ext: omni.kit.notification_manager-1.0.5] startup
[2.439s] [ext: omni.iray.libs-0.0.0] startup
[2.441s] [ext: omni.kit.window.filepicker-2.7.15] startup
[2.476s] [ext: omni.ui_query-1.1.1] startup
[2.477s] [ext: omni.mdl.neuraylib-0.1.0] startup
[2.479s] [ext: omni.kit.window.file_importer-1.0.10] startup
[2.479s] [ext: omni.kit.ui_test-1.2.9] startup
[2.481s] [ext: omni.mdl-0.1.0] startup
[2.494s] [ext: omni.kit.widget.zoombar-1.0.4] startup
[2.495s] [ext: omni.kit.widget.searchfield-1.0.10] startup
[2.496s] [ext: omni.kit.material.library-1.3.21] startup
[2.499s] [ext: omni.kit.browser.core-2.2.2] startup
[2.502s] [ext: omni.kit.window.file_exporter-1.0.10] startup
[2.503s] [ext: omni.physics.tensors-0.1.0] startup
[2.508s] [ext: omni.kit.browser.folder.core-1.7.3] startup
[2.511s] [ext: omni.kit.window.file-1.3.32] startup
[2.513s] [ext: omni.physx.tensors-0.1.0] startup
[2.518s] [ext: omni.isaac.version-1.0.0] startup
[2.518s] [ext: omni.kit.browser.sample-1.2.5] startup
[2.520s] [ext: omni.isaac.cloner-0.4.1] startup
[2.521s] [ext: omni.isaac.core-1.46.3] startup
[2.615s] [ext: omni.warp-0.6.3] startup
Warp 0.6.3 initialized:
CUDA Toolkit: 11.5, Driver: 12.1
Devices:
"cpu" | x86_64
"cuda:0" | NVIDIA GeForce RTX 4090 (sm_89)
"cuda:1" | NVIDIA GeForce RTX 4090 (sm_89)
Kernel cache: /home/bizon/.cache/warp/0.6.3
[2.860s] [ext: omni.kit.window.title-1.1.2] startup
[2.861s] [ext: omni.isaac.gym-0.3.3] startup
[2.861s] [ext: omni.isaac.sim.python.gym.headless-2022.2.1] startup
[2.861s] Simulation App Starting
2023-05-12 20:43:04 [3,009ms] [Warning] [omni.kvdb.plugin] repair failed
[3.017s] [ext: omni.kit.numpy.common-0.1.0] startup
[3.018s] [ext: omni.graph.nodes-1.48.3] startup
[3.031s] [ext: omni.isaac.dynamic_control-1.2.3] startup
[3.040s] [ext: omni.isaac.kit-1.4.1] startup
[3.040s] [ext: omni.kit.widget.path_field-2.0.4] startup
[3.041s] [ext: omni.kit.search_core-1.0.2] startup
[3.042s] [ext: omni.kit.widget.browser_bar-2.0.5] startup
[3.042s] [ext: omni.kit.widget.filebrowser-2.3.10] startup
[3.050s] [ext: omni.kit.widget.versioning-1.3.8] startup
[3.052s] [ext: omni.kit.notification_manager-1.0.5] startup
[3.053s] [ext: omni.iray.libs-0.0.0] startup
[3.056s] [ext: omni.kit.window.filepicker-2.7.15] startup
[3.092s] [ext: omni.ui_query-1.1.1] startup
[3.093s] [ext: omni.mdl.neuraylib-0.1.0] startup
[3.094s] [ext: omni.kit.window.file_importer-1.0.10] startup
[3.095s] [ext: omni.kit.ui_test-1.2.9] startup
[3.097s] [ext: omni.mdl-0.1.0] startup
[3.110s] [ext: omni.kit.widget.zoombar-1.0.4] startup
[3.111s] [ext: omni.kit.widget.searchfield-1.0.10] startup
[3.111s] [ext: omni.kit.material.library-1.3.21] startup
[3.115s] [ext: omni.kit.browser.core-2.2.2] startup
[3.118s] [ext: omni.kit.window.file_exporter-1.0.10] startup
[3.119s] [ext: omni.physics.tensors-0.1.0] startup
[3.125s] [ext: omni.kit.browser.folder.core-1.7.3] startup
[3.127s] [ext: omni.kit.window.file-1.3.32] startup
[3.129s] [ext: omni.physx.tensors-0.1.0] startup
[3.134s] [ext: omni.isaac.version-1.0.0] startup
[3.135s] [ext: omni.kit.browser.sample-1.2.5] startup
[3.136s] [ext: omni.isaac.cloner-0.4.1] startup
[3.137s] [ext: omni.isaac.core-1.46.3] startup
[3.232s] [ext: omni.warp-0.6.3] startup
Warp 0.6.3 initialized:
CUDA Toolkit: 11.5, Driver: 12.1
Devices:
"cpu" | x86_64
"cuda:0" | NVIDIA GeForce RTX 4090 (sm_89)
"cuda:1" | NVIDIA GeForce RTX 4090 (sm_89)
Kernel cache: /home/bizon/.cache/warp/0.6.3
[3.279s] app ready
2023-05-12 20:43:04 [3,272ms] [Warning] [omni.kit.browser.folder.core.models.folder_browser_model] Do not load cache for Warp because url changed:
2023-05-12 20:43:04 [3,272ms] [Warning] [omni.kit.browser.folder.core.models.folder_browser_model] - from /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.warp-0.6.1+cp37/data/scenes
2023-05-12 20:43:04 [3,272ms] [Warning] [omni.kit.browser.folder.core.models.folder_browser_model] - to /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.warp-0.6.3+cp37/data/scenes
[3.292s] Simulation App Startup Complete
task:
name: AnymalTerrain
physics_engine: physx
env:
numEnvs: 2048
numObservations: 188
numActions: 12
envSpacing: 3.0
terrain:
staticFriction: 1.0
dynamicFriction: 1.0
restitution: 0.0
curriculum: True
maxInitMapLevel: 0
mapLength: 8.0
mapWidth: 8.0
numLevels: 10
numTerrains: 20
terrainProportions: [0.1, 0.1, 0.35, 0.25, 0.2]
slopeTreshold: 0.5
baseInitState:
pos: [0.0, 0.0, 0.62]
rot: [1.0, 0.0, 0.0, 0.0]
vLinear: [0.0, 0.0, 0.0]
vAngular: [0.0, 0.0, 0.0]
randomCommandVelocityRanges:
linear_x: [-1.0, 1.0]
linear_y: [-1.0, 1.0]
yaw: [-3.14, 3.14]
control:
stiffness: 80.0
damping: 2.0
actionScale: 0.5
decimation: 4
defaultJointAngles:
LF_HAA: 0.03
LH_HAA: 0.03
RF_HAA: -0.03
RH_HAA: -0.03
LF_HFE: 0.4
LH_HFE: -0.4
RF_HFE: 0.4
RH_HFE: -0.4
LF_KFE: -0.8
LH_KFE: 0.8
RF_KFE: -0.8
RH_KFE: 0.8
learn:
terminalReward: 0.0
linearVelocityXYRewardScale: 1.0
linearVelocityZRewardScale: -4.0
angularVelocityXYRewardScale: -0.05
angularVelocityZRewardScale: 0.5
orientationRewardScale: -0.0
torqueRewardScale: -2e-05
jointAccRewardScale: -0.0005
baseHeightRewardScale: -0.0
actionRateRewardScale: -0.01
fallenOverRewardScale: -1.0
hipRewardScale: -0.0
linearVelocityScale: 2.0
angularVelocityScale: 0.25
dofPositionScale: 1.0
dofVelocityScale: 0.05
heightMeasurementScale: 5.0
addNoise: True
noiseLevel: 1.0
dofPositionNoise: 0.01
dofVelocityNoise: 1.5
linearVelocityNoise: 0.1
angularVelocityNoise: 0.2
gravityNoise: 0.05
heightMeasurementNoise: 0.06
pushInterval_s: 15
episodeLength_s: 20
sim:
dt: 0.005
use_gpu_pipeline: True
gravity: [0.0, 0.0, -9.81]
add_ground_plane: False
use_flatcache: True
enable_scene_query_support: False
disable_contact_processing: True
enable_cameras: False
default_physics_material:
static_friction: 1.0
dynamic_friction: 1.0
restitution: 0.0
physx:
worker_thread_count: 4
solver_type: 1
use_gpu: True
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
contact_offset: 0.02
rest_offset: 0.0
bounce_threshold_velocity: 0.2
friction_offset_threshold: 0.04
friction_correlation_distance: 0.025
enable_sleeping: True
enable_stabilization: True
max_depenetration_velocity: 100.0
gpu_max_rigid_contact_count: 524288
gpu_max_rigid_patch_count: 163840
gpu_found_lost_pairs_capacity: 4194304
gpu_found_lost_aggregate_pairs_capacity: 33554432
gpu_total_aggregate_pairs_capacity: 4194304
gpu_max_soft_body_contacts: 1048576
gpu_max_particle_contacts: 1048576
gpu_heap_capacity: 134217728
gpu_temp_buffer_capacity: 33554432
gpu_max_num_partitions: 8
anymal:
override_usd_defaults: False
enable_self_collisions: True
enable_gyroscopic_forces: False
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
sleep_threshold: 0.005
stabilization_threshold: 0.001
density: -1
max_depenetration_velocity: 100.0
train:
params:
seed: 42
algo:
name: a2c_continuous
model:
name: continuous_a2c_logstd
network:
name: actor_critic
separate: True
space:
continuous:
mu_activation: None
sigma_activation: None
mu_init:
name: default
sigma_init:
name: const_initializer
val: 0.0
fixed_sigma: True
mlp:
units: [512, 256, 128]
activation: elu
d2rl: False
initializer:
name: default
regularizer:
name: None
load_checkpoint: False
load_path:
config:
name: AnymalTerrain
full_experiment_name: AnymalTerrain
device: cuda:0
device_name: cuda:0
env_name: rlgpu
multi_gpu: True
ppo: True
mixed_precision: False
normalize_input: True
normalize_value: True
normalize_advantage: True
value_bootstrap: True
clip_actions: False
num_actors: 2048
reward_shaper:
scale_value: 1.0
gamma: 0.99
tau: 0.95
e_clip: 0.2
entropy_coef: 0.001
learning_rate: 0.0003
lr_schedule: adaptive
kl_threshold: 0.008
truncate_grads: True
grad_norm: 1.0
horizon_length: 48
minibatch_size: 16384
mini_epochs: 5
critic_coef: 2
clip_value: True
seq_len: 4
bounds_loss_coef: 0.0
max_epochs: 2000
save_best_after: 100
score_to_win: 20000
save_frequency: 50
print_stats: True
task_name: AnymalTerrain
experiment:
num_envs:
seed: 42
torch_deterministic: False
max_iterations:
physics_engine: physx
pipeline: gpu
sim_device: gpu
device_id: 0
rl_device: cuda:0
multi_gpu: True
num_threads: 4
solver_type: 1
test: False
checkpoint:
headless: True
enable_livestream: False
mt_timeout: 30
wandb_activate: False
wandb_group:
wandb_name: AnymalTerrain
wandb_entity:
wandb_project: omniisaacgymenvs
Setting seed: 42
Sim params does not have attribute: physx
Sim params does not have attribute: anymal
Pipeline: GPU
Pipeline Device: cuda:0
Sim Device: GPU
[3.314s] [ext: omni.inspect-1.0.1] startup
[3.320s] [ext: omni.kit.clipboard-1.0.0] startup
[3.322s] [ext: omni.kit.menu.create-1.0.8] startup
[3.328s] [ext: omni.volume-0.1.0] startup
[3.330s] [ext: omni.kit.context_menu-1.5.12] startup
[3.333s] [ext: omni.activity.core-1.0.1] startup
[3.335s] [ext: omni.hydra.rtx-0.1.0] startup
[3.341s] [ext: omni.debugdraw-0.1.1] startup
[3.345s] [ext: omni.kit.widget.stage-2.7.24] startup
[3.349s] [ext: omni.kit.window.property-1.8.2] startup
[3.351s] [ext: omni.kit.viewport.utility-1.0.14] startup
[3.351s] [ext: omni.kit.property.usd-3.18.17] startup
[3.358s] [ext: omni.kit.widget.text_editor-1.0.2] startup
[3.359s] [ext: omni.kit.widget.settings-1.0.1] startup
[3.360s] [ext: omni.kit.widget.graph-1.5.6-104_2] startup
[3.370s] [ext: omni.ui.scene-1.5.18] startup
[3.374s] [ext: omni.kit.window.preferences-1.3.8] startup
[3.405s] [ext: omni.kit.window.extensions-1.1.1] startup
[3.409s] [ext: omni.kit.widget.prompt-1.0.5] startup
[3.410s] [ext: omni.graph.ui-1.24.2] startup
[3.437s] [ext: omni.graph.scriptnode-0.10.0] startup
[3.439s] [ext: omni.graph.action-1.31.1] startup
[3.445s] [ext: omni.graph.bundle.action-1.3.0] startup
[3.445s] [ext: omni.syntheticdata-0.2.4] startup
2023-05-12 20:43:05 [3,440ms] [Warning] [omni.syntheticdata.scripts.extension] SyntheticData extension needs at least a stageFrameHistoryCount of 3
[3.458s] [ext: omni.command.usd-1.0.2] startup
[3.459s] [ext: omni.replicator.core-1.7.8] startup
[3.482s] [ext: omni.kit.window.title-1.1.2] startup
[3.482s] [ext: omni.isaac.gym-0.3.3] startup
[3.482s] [ext: omni.isaac.sim.python.gym.headless-2022.2.1] startup
[3.483s] Simulation App Starting
2023-05-12 20:43:05 [3,461ms] [Warning] [omni.replicator.core.scripts.annotators] Annotator PostProcessDispatch is already registered, overwriting annotator template
[3.513s] [ext: omni.replicator.isaac-1.7.4] startup
Task Device: cuda:0
RL device: cuda:0
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/gym/spaces/box.py:84: UserWarning: WARN: Box bound precision lowered by casting to float32
logger.warn(f"Box bound precision lowered by casting to {self.dtype}")
[3.863s] app ready
2023-05-12 20:43:05 [3,862ms] [Warning] [omni.kit.browser.folder.core.models.folder_browser_model] Do not load cache for Warp because url changed:
2023-05-12 20:43:05 [3,862ms] [Warning] [omni.kit.browser.folder.core.models.folder_browser_model] - from /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.warp-0.6.1+cp37/data/scenes
2023-05-12 20:43:05 [3,862ms] [Warning] [omni.kit.browser.folder.core.models.folder_browser_model] - to /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.warp-0.6.3+cp37/data/scenes
[3.876s] Simulation App Startup Complete
task:
name: AnymalTerrain
physics_engine: physx
env:
numEnvs: 2048
numObservations: 188
numActions: 12
envSpacing: 3.0
terrain:
staticFriction: 1.0
dynamicFriction: 1.0
restitution: 0.0
curriculum: True
maxInitMapLevel: 0
mapLength: 8.0
mapWidth: 8.0
numLevels: 10
numTerrains: 20
terrainProportions: [0.1, 0.1, 0.35, 0.25, 0.2]
slopeTreshold: 0.5
baseInitState:
pos: [0.0, 0.0, 0.62]
rot: [1.0, 0.0, 0.0, 0.0]
vLinear: [0.0, 0.0, 0.0]
vAngular: [0.0, 0.0, 0.0]
randomCommandVelocityRanges:
linear_x: [-1.0, 1.0]
linear_y: [-1.0, 1.0]
yaw: [-3.14, 3.14]
control:
stiffness: 80.0
damping: 2.0
actionScale: 0.5
decimation: 4
defaultJointAngles:
LF_HAA: 0.03
LH_HAA: 0.03
RF_HAA: -0.03
RH_HAA: -0.03
LF_HFE: 0.4
LH_HFE: -0.4
RF_HFE: 0.4
RH_HFE: -0.4
LF_KFE: -0.8
LH_KFE: 0.8
RF_KFE: -0.8
RH_KFE: 0.8
learn:
terminalReward: 0.0
linearVelocityXYRewardScale: 1.0
linearVelocityZRewardScale: -4.0
angularVelocityXYRewardScale: -0.05
angularVelocityZRewardScale: 0.5
orientationRewardScale: -0.0
torqueRewardScale: -2e-05
jointAccRewardScale: -0.0005
baseHeightRewardScale: -0.0
actionRateRewardScale: -0.01
fallenOverRewardScale: -1.0
hipRewardScale: -0.0
linearVelocityScale: 2.0
angularVelocityScale: 0.25
dofPositionScale: 1.0
dofVelocityScale: 0.05
heightMeasurementScale: 5.0
addNoise: True
noiseLevel: 1.0
dofPositionNoise: 0.01
dofVelocityNoise: 1.5
linearVelocityNoise: 0.1
angularVelocityNoise: 0.2
gravityNoise: 0.05
heightMeasurementNoise: 0.06
pushInterval_s: 15
episodeLength_s: 20
sim:
dt: 0.005
use_gpu_pipeline: True
gravity: [0.0, 0.0, -9.81]
add_ground_plane: False
use_flatcache: True
enable_scene_query_support: False
disable_contact_processing: True
enable_cameras: False
default_physics_material:
static_friction: 1.0
dynamic_friction: 1.0
restitution: 0.0
physx:
worker_thread_count: 4
solver_type: 1
use_gpu: True
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
contact_offset: 0.02
rest_offset: 0.0
bounce_threshold_velocity: 0.2
friction_offset_threshold: 0.04
friction_correlation_distance: 0.025
enable_sleeping: True
enable_stabilization: True
max_depenetration_velocity: 100.0
gpu_max_rigid_contact_count: 524288
gpu_max_rigid_patch_count: 163840
gpu_found_lost_pairs_capacity: 4194304
gpu_found_lost_aggregate_pairs_capacity: 33554432
gpu_total_aggregate_pairs_capacity: 4194304
gpu_max_soft_body_contacts: 1048576
gpu_max_particle_contacts: 1048576
gpu_heap_capacity: 134217728
gpu_temp_buffer_capacity: 33554432
gpu_max_num_partitions: 8
anymal:
override_usd_defaults: False
enable_self_collisions: True
enable_gyroscopic_forces: False
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
sleep_threshold: 0.005
stabilization_threshold: 0.001
density: -1
max_depenetration_velocity: 100.0
train:
params:
seed: 42
algo:
name: a2c_continuous
model:
name: continuous_a2c_logstd
network:
name: actor_critic
separate: True
space:
continuous:
mu_activation: None
sigma_activation: None
mu_init:
name: default
sigma_init:
name: const_initializer
val: 0.0
fixed_sigma: True
mlp:
units: [512, 256, 128]
activation: elu
d2rl: False
initializer:
name: default
regularizer:
name: None
load_checkpoint: False
load_path:
config:
name: AnymalTerrain
full_experiment_name: AnymalTerrain
device: cuda:1
device_name: cuda:1
env_name: rlgpu
multi_gpu: True
ppo: True
mixed_precision: False
normalize_input: True
normalize_value: True
normalize_advantage: True
value_bootstrap: True
clip_actions: False
num_actors: 2048
reward_shaper:
scale_value: 1.0
gamma: 0.99
tau: 0.95
e_clip: 0.2
entropy_coef: 0.001
learning_rate: 0.0003
lr_schedule: adaptive
kl_threshold: 0.008
truncate_grads: True
grad_norm: 1.0
horizon_length: 48
minibatch_size: 16384
mini_epochs: 5
critic_coef: 2
clip_value: True
seq_len: 4
bounds_loss_coef: 0.0
max_epochs: 2000
save_best_after: 100
score_to_win: 20000
save_frequency: 50
print_stats: True
task_name: AnymalTerrain
experiment:
num_envs:
seed: 42
torch_deterministic: False
max_iterations:
physics_engine: physx
pipeline: gpu
sim_device: gpu
device_id: 1
rl_device: cuda:1
multi_gpu: True
num_threads: 4
solver_type: 1
test: False
checkpoint:
headless: True
enable_livestream: False
mt_timeout: 30
wandb_activate: False
wandb_group:
wandb_name: AnymalTerrain
wandb_entity:
wandb_project: omniisaacgymenvs
Setting seed: 42
Sim params does not have attribute: physx
Sim params does not have attribute: anymal
Pipeline: GPU
Pipeline Device: cuda:1
Sim Device: GPU
[3.898s] [ext: omni.inspect-1.0.1] startup
[3.904s] [ext: omni.kit.clipboard-1.0.0] startup
[3.907s] [ext: omni.kit.menu.create-1.0.8] startup
[3.913s] [ext: omni.volume-0.1.0] startup
[3.915s] [ext: omni.kit.context_menu-1.5.12] startup
[3.918s] [ext: omni.activity.core-1.0.1] startup
[3.920s] [ext: omni.hydra.rtx-0.1.0] startup
[3.926s] [ext: omni.debugdraw-0.1.1] startup
[3.930s] [ext: omni.kit.widget.stage-2.7.24] startup
[3.934s] [ext: omni.kit.window.property-1.8.2] startup
[3.936s] [ext: omni.kit.viewport.utility-1.0.14] startup
[3.937s] [ext: omni.kit.property.usd-3.18.17] startup
[3.943s] [ext: omni.kit.widget.text_editor-1.0.2] startup
[3.944s] [ext: omni.kit.widget.settings-1.0.1] startup
[3.945s] [ext: omni.kit.widget.graph-1.5.6-104_2] startup
[3.955s] [ext: omni.ui.scene-1.5.18] startup
[3.960s] [ext: omni.kit.window.preferences-1.3.8] startup
[3.991s] [ext: omni.kit.window.extensions-1.1.1] startup
[3.995s] [ext: omni.kit.widget.prompt-1.0.5] startup
[3.996s] [ext: omni.graph.ui-1.24.2] startup
[4.024s] [ext: omni.graph.scriptnode-0.10.0] startup
[4.026s] [ext: omni.graph.action-1.31.1] startup
[4.032s] [ext: omni.graph.bundle.action-1.3.0] startup
[4.033s] [ext: omni.syntheticdata-0.2.4] startup
2023-05-12 20:43:05 [4,034ms] [Warning] [omni.syntheticdata.scripts.extension] SyntheticData extension needs at least a stageFrameHistoryCount of 3
[4.045s] [ext: omni.command.usd-1.0.2] startup
[4.046s] [ext: omni.replicator.core-1.7.8] startup
2023-05-12 20:43:05 [4,055ms] [Warning] [omni.replicator.core.scripts.annotators] Annotator PostProcessDispatch is already registered, overwriting annotator template
[4.101s] [ext: omni.replicator.isaac-1.7.4] startup
Task Device: cuda:1
RL device: cuda:1
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/gym/spaces/box.py:84: UserWarning: WARN: Box bound precision lowered by casting to float32
logger.warn(f"Box bound precision lowered by casting to {self.dtype}")
/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3190.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
2023-05-12 20:43:05 [4,357ms] [Warning] [omni.isaac.core.utils.viewports] could not get active viewport, cannot set camera view
/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/functional.py:504: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3190.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
2023-05-12 20:43:07 [5,553ms] [Warning] [omni.isaac.core.utils.viewports] could not get active viewport, cannot set camera view
self.seed = 42
Started to train
Exact experiment name requested from command line: AnymalTerrain
self.seed = 43
Started to train
Exact experiment name requested from command line: AnymalTerrain
Box(-1.0, 1.0, (12,), float32) Box(-inf, inf, (188,), float32)
current training device: cuda:1
build mlp: 188
build mlp: 188
RunningMeanStd: (1,)
RunningMeanStd: (188,)
Box(-1.0, 1.0, (12,), float32) Box(-inf, inf, (188,), float32)
current training device: cuda:0
[2023-05-12 10:43:25] Running RL reset
build mlp: 188
build mlp: 188
RunningMeanStd: (1,)
RunningMeanStd: (188,)
[2023-05-12 10:43:25] Running RL reset
====================broadcasting parameters
====================broadcasting parameters
[E ProcessGroupNCCL.cpp:821] [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=1, OpType=BROADCAST, Timeout(ms)=1800000) ran for 1806521 milliseconds before timing out.
Error executing job with overrides: ['headless=True', 'task=AnymalTerrain', 'multi_gpu=True']
[E ProcessGroupNCCL.cpp:456] Some NCCL operations have failed or timed out. Due to the asynchronous nature of CUDA kernels, subsequent GPU operations might run on corrupted/incomplete data.
[E ProcessGroupNCCL.cpp:461] To avoid data inconsistency, we are taking the entire process down.
terminate called after throwing an instance of 'std::runtime_error'
what(): [Rank 1] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=1, OpType=BROADCAST, Timeout(ms)=1800000) ran for 1806521 milliseconds before timing out.
WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 28378 closing signal SIGTERM
[E ProcessGroupNCCL.cpp:821] [Rank 0] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=3, OpType=ALLREDUCE, Timeout(ms)=1800000) ran for 1806882 milliseconds before timing out.
[E ProcessGroupNCCL.cpp:456] Some NCCL operations have failed or timed out. Due to the asynchronous nature of CUDA kernels, subsequent GPU operations might run on corrupted/incomplete data.
[E ProcessGroupNCCL.cpp:461] To avoid data inconsistency, we are taking the entire process down.
terminate called after throwing an instance of 'std::runtime_error'
what(): [Rank 0] Watchdog caught collective operation timeout: WorkNCCL(SeqNum=3, OpType=ALLREDUCE, Timeout(ms)=1800000) ran for 1806882 milliseconds before timing out.
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -6) local_rank: 1 (pid: 28379) of binary: /home/bizon/anaconda3/envs/isaac-sim/bin/python
Traceback (most recent call last):
File "/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/runpy.py", line 193, in _run_module_as_main
"main", mod_spec)
File "/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/distributed/run.py", line 766, in
main()
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/distributed/elastic/multiprocessing/errors/init.py", line 346, in wrapper
return f(*args, **kwargs)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/distributed/run.py", line 762, in main
run(args)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/distributed/run.py", line 756, in run
)(*cmd_args)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/distributed/launcher/api.py", line 132, in call
return launch_agent(self._config, self._entrypoint, list(args))
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.1/extscache/omni.pip.torch-1_13_1-0.1.4+104.2.lx64/torch-1-13-1/torch/distributed/launcher/api.py", line 248, in launch_agent
failures=result.failures,
torch.distributed.elastic.multiprocessing.errors.ChildFailedError:

scripts/rlgames_train.py FAILED

Failures:
<NO_OTHER_FAILURES>

Root Cause (first observed failure):
[0]:
time : 2023-05-12_11:13:37
host : dl
rank : 1 (local_rank: 1)
exitcode : -6 (pid: 28379)
error_file: <N/A>
traceback : Signal 6 (SIGABRT) received by PID 28379

[Bug report] Shadow hand interpenetration + Missing knuckle visual

Hi there,

1) Interpenetration

I don't know if this is a bug or not, but I faced an interpenetration issue while manipulating the Shadow Hand model. As you can see in the screenshot below, fingers sometimes interpenetrate each other, as well as the palm. I'd tend to say that this behavior is not wanted (maybe naively), since such interpenetration doesn't happen with a physical robot.

I looked in the config file of the task and the sim.physx.shadow_hand.enable_self_collisions parameter was set to True. Setting it to False didn't fix the issue. Could you provide me with some info/workaround about that?

2) Missing visual

When inspecting the shadow_hand_instanceable.usd asset, I noticed that the first finger knuckle mesh seems to be missing (as visible below). This is not critical, but if that's indeed the case, I thought it would be worth mentioning it here(?) . Note that I fixed the issue in my local usd file.

How does Randomization on Material Properties Work?

Hi everyone,

I want to domain randomize the dynamic and static friction coefficients between the agent and the ground. I follow the instruction (modified the config file, added codes needed for DR), but it seams that the domain randomization does not work.

Here's how I check if it works or not:

1. I create a rigid_prim_view for the meshes whose material properties should be changed.
2. I set the randomization parameters as follows:
   rigid_prim_views:
   knee_view:
   material_properties:
   on_reset:
   num_buckets: 250
   operation: "scaling"
   distribution: "uniform"
   distribution_parameters: [[0.0, 0.0, 1], [0.05, 0.05, 1]]
3. I start up the simulation. According to the parameters I set, the friction between the agent and the ground should be very low, and the ground should be very "slippery". However, the agent still walks very normally.

So I want to know what is the correct way to randomize the material properties? Do I need to first assign a material to the meshes? Actually I tried to create and assign materials to the meshes, but when I started training, the friction parameters in the UI did not change at all (I enabled the interaction between the env and the UI);

And I noticed that in the config file, there's a parameter called "default_physics_material". I suppose if I did not create any material for my agent, this will be the material for every mesh. Is this correct?

[Error] [carb.events.python] AttributeError: 'PlayButtonGroup' object has no attribute '_play_button'

Hi,

when I set enable_cameras=True and try to run a training in headless mode ($ isaacpython omniisaacgymenvs/scripts/rlgames_train.py task=AllegroHand num_envs=4 test=True headless=True), I get this error

[Error] [carb.events.python] AttributeError: 'PlayButtonGroup' object has no attribute '_play_button'
At:
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/exts/omni.kit.widget.toolbar/omni/kit/widget/toolbar/builtin_tools/play_button_group.py(129): _on_timeline_event
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.isaac.core/omni/isaac/core/simulation_context/simulation_context.py(549): play
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.isaac.core/omni/isaac/core/simulation_context/simulation_context.py(387): initialize_physics
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.isaac.core/omni/isaac/core/simulation_context/simulation_context.py(408): reset
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.isaac.core/omni/isaac/core/world/world.py(281): reset
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/exts/omni.isaac.gym/omni/isaac/gym/vec_env/vec_env_base.py(94): set_task
  /home/johannes/work/isaac-sim-envs/omniisaacgymenvs/envs/vec_env_rlgames.py(51): set_task
  /home/johannes/work/isaac-sim-envs/omniisaacgymenvs/utils/task_util.py(72): initialize_task
  omniisaacgymenvs/scripts/rlgames_train.py(114): parse_hydra_configs
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/core/utils.py(160): run_job
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/hydra.py(106): run
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py(381): <lambda>
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py(211): run_and_report
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py(378): _run_hydra
  /home/johannes/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/main.py(52): decorated_main
  omniisaacgymenvs/scripts/rlgames_train.py(143): <module>

The simulation continues to run and using something like this we can even get camera images, but I would still like to resolve the error to make sure we don't run into problem later on.

       cam = rep.create.camera(
            position=position, look_at=look_at, clipping_range=(0.01, 10)
        )
        RESOLUTION = self._cfg["video_resolution"]
        self._render_product_path = rep.create.render_product(
            cam, resolution=(RESOLUTION, RESOLUTION)
        )
        self.rgb = rep.AnnotatorRegistry.get_annotator("rgb")
        self.rgb.attach([self._render_product_path])
        self.rgb.get_data()

[Question] Adding contact sensors to shadow hand's fingertips

Hi there,

Context

As stated here, I'm currently modifying the ShadowHand example you provide in order to control the hand using efforts rather than position control, and I'd like to train RL models to achieve this using contact/force/tactile observations rather than "visual" ones (fingertip positions, angles, velocities, etc.) as done in the OpenAI examples for instance.

Wanted

In order to simulate the pressure/tactile sensors of the real Shadow Hand, I want to attach a contact sensor (as defined here) to all fingertips of all envs, in order to get contact data at each step of my RL loop.

Tried

In the set_up_scenemethod of my task class, I tried to create such sensors using the IsaacSensorCreateContactSensor command with self.default_zero_env_path + "/shadow_hand/robot0_xxdistal" as the prim path (xx being replaced by ff, mf, ...), and I then used the ContactSensorInterface in get_observations to get data from these sensors. I observed nothing but zero measurements, so I tried to visualize my sensors in the simulator by setting the visualize flag to True when using the creation command, and I set the color parameter to some kind of red. Turns out the sensors are located way above the hand (in the picture below, see the red dots outlining the spherical sensors), and do not move although the fingers do, which is not wanted.

When setting use_flatcache to False in the config file, the hand appears standing vertically, and does not move anymore. However, the sensors' location matches in this case the position of the fingertips of this vertically standing hand.

I also tried to create a ContactSensor object without the mentioned command, setting the prim path as stated before, and adding the sensor to the scene, but this was unsuccessful.

Could someone help me creating contact sensors attached to the controlled fingers? Thanks very much in advance!

Get RGB and Depth Data in headless mode

Is it possible to read out the depth information for each camera (mounted at the end effector of each manipulator) in headless mode?

I'm currently training an RL policy with 4096 robots in parallel. By activating the visualization (headless=False), the training process takes significantly longer.

[Question] Effort control mode (instead of position control)

Hi there,

I'm currently modifying the ShadowHand example you provide in order to control the hand using efforts rather than position control. To do that, I use the set_joint_efforts method on the hand view (instance of ShadowHandView(ArticulationView)) instead of setting position targets and using set_joint_position_targets as done in the provided example.

This seems to work (the hand does move when using a random policy), but since the get_applied_joint_efforts method isn't working (apparently a known bug that will be fixed soon, or is fixed and the problem is on my side) and returns nothing but zero efforts in the joints, I cannot really investigate and check that everything is working properly.

What I am curious about is the set_drive call in the ShadowHand class. This call sets the drive type to either linear or angular through pxr.UsdPhysics.DriveAPI, as well as the target position or velocity to 0.0. Is this some kind of initialization step required with usd, or is this related to some behind-the-scene mechanism underlying the control of the hand, not depending on the type of control I choose using the Python API (e.g. by using switch_control_mode), and preventing me from doing effort control on the hand?

I apologize if this is a silly question, and don't hesitate to tell me if it shoud be asked on the dev forum rather than here. Thanks!

[Error: Segmentation fault $python_exe "$@" $args], When i run the change parameter fixed_base.

Hi all,
i want to perform my new RL task and the new RL task was modified from the example cartpole. I change the default parameter from [fixed_base = false ] to [fixed_base = true ] in the Cartpole.yaml , as a result ,the segmentation fault happend. And i test the problem in oher examples, such as Ant and Allegrohand, the same error occured.
however, i used to change the fixed_base parameter in former isaacgym version [ isaacgym_preview3] which was independent of isaac sim, and no error happend.

following is the error list:
Cartpole:
override_usd_defaults: False
fixed_base: True // i change the default value
enable_self_collisions: False
enable_gyroscopic_forces: True
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
sleep_threshold: 0.005
stabilization_threshold: 0.001
density: -1
max_depenetration_velocity: 100.0
contact_offset: 0.02
rest_offset: 0.001

.local/share/ov/pkg/isaac_sim-2022.1.0/python.sh: line 46: 20430 Segmentation fault $python_exe "$@" $args
There was an error running python

Process finished with exit code 1

Appreciate it if you can give some hints.
mao

running torch.distributed.run on my dual 4090 GPU workstation crashes immediately when training the Ant example.

# I am using a conda environment which has been set up properly as per instructions
python -m torch.distributed.run --nnodes=1 --nproc_per_node=2 scripts/rlgames_train.py headless=True task=Ant multi_gpu=True

Any ideas? I would love to be able to use my other 4090!

stack trace:

python -m torch.distributed.run --nnodes=1 --nproc_per_node=2 scripts/rlgames_train.py headless=True task=Ant multi_gpu=True
WARNING:main:

---

Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed.

---

/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config': Defaults list is missing _self_. See https://hydra.cc/docs/upgrades/1.0_to_1.1/default_composition_order for more information
warnings.warn(msg, UserWarning)
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:412: UserWarning: In config: Invalid overriding of hydra/job_logging:
Default list overrides requires 'override' keyword.
See https://hydra.cc/docs/next/upgrades/1.0_to_1.1/defaults_list_override for more information.

deprecation_warning(msg)
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config': Defaults list is missing _self_. See https://hydra.cc/docs/upgrades/1.0_to_1.1/default_composition_order for more information
warnings.warn(msg, UserWarning)
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:412: UserWarning: In config: Invalid overriding of hydra/job_logging:
Default list overrides requires 'override' keyword.
See https://hydra.cc/docs/next/upgrades/1.0_to_1.1/defaults_list_override for more information.

deprecation_warning(msg)
[Warning] [omni.isaac.kit.simulation_app] Modules: ['omniisaacgymenvs', 'omniisaacgymenvs.utils', 'omniisaacgymenvs.utils.hydra_cfg', 'omniisaacgymenvs.utils.hydra_cfg.hydra_utils', 'omniisaacgymenvs.utils.hydra_cfg.reformat', 'omniisaacgymenvs.utils.rlgames', 'omniisaacgymenvs.utils.rlgames.rlgames_utils', 'omniisaacgymenvs.utils.task_util', 'omniisaacgymenvs.utils.config_utils', 'omniisaacgymenvs.utils.config_utils.path_utils', 'omniisaacgymenvs.envs', 'omniisaacgymenvs.envs.vec_env_rlgames', 'omni', 'omni.isaac.gym', 'omni.isaac.gym.vec_env', 'omni.isaac.gym.vec_env.vec_env_base', 'omni.kit.app._impl.telemetry_helpers', 'omni.isaac.gym.vec_env.vec_env_mt'] were loaded before SimulationApp was started and might not be loaded correctly.
[Warning] [omni.isaac.kit.simulation_app] Please check to make sure no extra omniverse or pxr modules are imported before the call to SimulationApp(...)
Starting kit application with the following args: ['/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.isaac.kit/omni/isaac/kit/simulation_app.py', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/apps/omni.isaac.sim.python.gym.headless.kit', '--/app/tokens/exe-path=/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/kit', '--/persistent/app/viewport/displayOptions=3094', '--/rtx/materialDb/syncLoads=True', '--/rtx/hydra/materialSyncLoads=True--/omni.kit.plugin/syncUsdLoads=True', '--/app/renderer/resolution/width=1280', '--/app/renderer/resolution/height=720', '--/app/window/width=1440', '--/app/window/height=900', '--/renderer/multiGpu/enabled=True', '--/app/fastShutdown=True', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/apps', '--/physics/cudaDevice=0', '--portable', '--no-window']
Passing the following args to the base kit application: ['headless=True', 'task=Ant', 'multi_gpu=True']
[Warning] [omni.isaac.kit.simulation_app] Modules: ['omniisaacgymenvs', 'omniisaacgymenvs.utils', 'omniisaacgymenvs.utils.hydra_cfg', 'omniisaacgymenvs.utils.hydra_cfg.hydra_utils', 'omniisaacgymenvs.utils.hydra_cfg.reformat', 'omniisaacgymenvs.utils.rlgames', 'omniisaacgymenvs.utils.rlgames.rlgames_utils', 'omniisaacgymenvs.utils.task_util', 'omniisaacgymenvs.utils.config_utils', 'omniisaacgymenvs.utils.config_utils.path_utils', 'omniisaacgymenvs.envs', 'omniisaacgymenvs.envs.vec_env_rlgames', 'omni', 'omni.isaac.gym', 'omni.isaac.gym.vec_env', 'omni.isaac.gym.vec_env.vec_env_base', 'omni.kit.app._impl.telemetry_helpers', 'omni.isaac.gym.vec_env.vec_env_mt'] were loaded before SimulationApp was started and might not be loaded correctly.
[Warning] [omni.isaac.kit.simulation_app] Please check to make sure no extra omniverse or pxr modules are imported before the call to SimulationApp(...)
Starting kit application with the following args: ['/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.isaac.kit/omni/isaac/kit/simulation_app.py', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/apps/omni.isaac.sim.python.gym.headless.kit', '--/app/tokens/exe-path=/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/kit', '--/persistent/app/viewport/displayOptions=3094', '--/rtx/materialDb/syncLoads=True', '--/rtx/hydra/materialSyncLoads=True--/omni.kit.plugin/syncUsdLoads=True', '--/app/renderer/resolution/width=1280', '--/app/renderer/resolution/height=720', '--/app/window/width=1440', '--/app/window/height=900', '--/renderer/multiGpu/enabled=True', '--/app/fastShutdown=True', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts', '--ext-folder', '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/apps', '--/physics/cudaDevice=0', '--portable', '--no-window']
Passing the following args to the base kit application: ['headless=True', 'task=Ant', 'multi_gpu=True']
[Warning] [omni.kit.app.plugin] No crash reporter present, dumps uploading isn't available.
[Info] [carb] Logging to file: /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/kit/logs/Kit/Isaac-Sim/2022.2/kit_20230504_201332.log
2023-05-05 06:13:32 [1ms] [Warning] [omni.ext.plugin] [ext: omni.drivesim.sensors.nv.radar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.radar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.radar/config'
2023-05-05 06:13:32 [1ms] [Warning] [omni.ext.plugin] [ext: omni.drivesim.sensors.nv.lidar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.lidar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.lidar/config'
[Warning] [omni.kit.app.plugin] No crash reporter present, dumps uploading isn't available.
[Info] [carb] Logging to file: /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/kit/logs/Kit/Isaac-Sim/2022.2/kit_20230504_201332.log
2023-05-05 06:13:32 [1ms] [Warning] [omni.ext.plugin] [ext: omni.drivesim.sensors.nv.lidar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.lidar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.lidar/config'
2023-05-05 06:13:32 [1ms] [Warning] [omni.ext.plugin] [ext: omni.drivesim.sensors.nv.radar] Extensions config 'extension.toml' doesn't exist '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.radar' or '/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.drivesim.sensors.nv.radar/config'
[0.090s] [ext: omni.stats-0.0.0] startup
[0.090s] [ext: omni.stats-0.0.0] startup
[0.094s] [ext: omni.rtx.shadercache-1.0.0] startup
[0.094s] [ext: omni.rtx.shadercache-1.0.0] startup
[0.098s] [ext: omni.assets.plugins-0.0.0] startup
[0.099s] [ext: omni.gpu_foundation-0.0.0] startup
[0.097s] [ext: omni.assets.plugins-0.0.0] startup
[0.098s] [ext: omni.gpu_foundation-0.0.0] startup
2023-05-05 06:13:32 [96ms] [Warning] [carb] FrameworkImpl::setDefaultPlugin(client: omni.gpu_foundation_factory.plugin, desc : [carb::graphics::Graphics v2.11], plugin : carb.graphics-vulkan.plugin) failed. Plugin selection is locked, because the interface was previously acquired by:
2023-05-05 06:13:32 [95ms] [Warning] [carb] FrameworkImpl::setDefaultPlugin(client: omni.gpu_foundation_factory.plugin, desc : [carb::graphics::Graphics v2.11], plugin : carb.graphics-vulkan.plugin) failed. Plugin selection is locked, because the interface was previously acquired by:
[0.106s] [ext: carb.windowing.plugins-1.0.0] startup
[0.105s] [ext: carb.windowing.plugins-1.0.0] startup
2023-05-05 06:13:32 [105ms] [Warning] [carb.windowing-glfw.gamepad] Joystick with unknown remapping detected (will be ignored): ASRock LED Controller [03000000ce260000a201000010010000]
[0.113s] [ext: omni.kit.renderer.init-0.0.0] startup
2023-05-05 06:13:32 [105ms] [Warning] [carb.windowing-glfw.gamepad] Joystick with unknown remapping detected (will be ignored): ASRock LED Controller [03000000ce260000a201000010010000]
[0.112s] [ext: omni.kit.renderer.init-0.0.0] startup

|---------------------------------------------------------------------------------------------|
| Driver Version: 530.30.2 | Graphics API: Vulkan
|=============================================================================================|
| GPU | Name | Active | LDA | GPU Memory | Vendor-ID | LUID |
| | | | | | Device-ID | UUID |
|---------------------------------------------------------------------------------------------|
| 0 | NVIDIA GeForce RTX 4090 | Yes: 1 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 527cadd8.. |
|---------------------------------------------------------------------------------------------|
| 1 | NVIDIA GeForce RTX 4090 | Yes: 0 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 7c9328c9.. |
|=============================================================================================|
| OS: Linux dl, Version: 5.15.0-69-generic
| XServer Vendor: The X.Org Foundation, XServer Version: 12013000 (1.20.13.0)
| Processor: AMD Ryzen Threadripper PRO 5975WX 32-Cores | Cores: Unknown | Logical: 64
|---------------------------------------------------------------------------------------------|
| Total Memory (MB): 257558 | Free Memory: 230801
| Total Page/Swap (MB): 2047 | Free Page/Swap: 2047
|---------------------------------------------------------------------------------------------|

|---------------------------------------------------------------------------------------------|
| Driver Version: 530.30.2 | Graphics API: Vulkan
|=============================================================================================|
| GPU | Name | Active | LDA | GPU Memory | Vendor-ID | LUID |
| | | | | | Device-ID | UUID |
|---------------------------------------------------------------------------------------------|
| 0 | NVIDIA GeForce RTX 4090 | Yes: 1 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 527cadd8.. |
|---------------------------------------------------------------------------------------------|
| 1 | NVIDIA GeForce RTX 4090 | Yes: 0 | | 24810 MB | 10de | 0 |
| | | | | | 2684 | 7c9328c9.. |
|=============================================================================================|
| OS: Linux dl, Version: 5.15.0-69-generic
| XServer Vendor: The X.Org Foundation, XServer Version: 12013000 (1.20.13.0)
| Processor: AMD Ryzen Threadripper PRO 5975WX 32-Cores | Cores: Unknown | Logical: 64
|---------------------------------------------------------------------------------------------|
| Total Memory (MB): 257558 | Free Memory: 230801
| Total Page/Swap (MB): 2047 | Free Page/Swap: 2047
|---------------------------------------------------------------------------------------------|
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] IOMMU is enabled. Found 50 items in /sys/kernel/iommu_groups/.
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] IOMMU is enabled. Found 50 items in /sys/kernel/iommu_groups/.
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] On Linux only, CUDA and the display driver does not support IOMMU-enabled bare-metal PCIe peer to peer memory copy.
However, CUDA and the display driver does support IOMMU via VM pass through. As a consequence, users on Linux,
when running on a native bare metal system, should disable the IOMMU. The IOMMU should be enabled and the VFIO driver
be used as a PCIe pass through for virtual machines.
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] On Linux only, CUDA and the display driver does not support IOMMU-enabled bare-metal PCIe peer to peer memory copy.
However, CUDA and the display driver does support IOMMU via VM pass through. As a consequence, users on Linux,
when running on a native bare metal system, should disable the IOMMU. The IOMMU should be enabled and the VFIO driver
be used as a PCIe pass through for virtual machines.
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin]
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin]
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] !!!!! Local system validation failed! Incorrect configuration detected.
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] Summary below. Details above.
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] !!!!! Local system validation failed! Incorrect configuration detected.
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] Summary below. Details above.
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin]
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] - ECC: OK
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] - IOMMU: FAILED
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin]
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin]
2023-05-05 06:13:33 [875ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] - ECC: OK
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] - IOMMU: FAILED
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin]
2023-05-05 06:13:33 [873ms] [Warning] [gpu.foundation.plugin] -----------------------------------------------------------------------
[1.630s] [ext: omni.kit.pipapi-0.0.0] startup
[1.634s] [ext: omni.kit.pip_archive-0.0.0] startup
[1.641s] [ext: omni.kit.loop-isaac-1.0.0] startup
[1.641s] [ext: omni.kit.async_engine-0.0.0] startup
[1.643s] [ext: omni.kit.test-0.0.0] startup
[1.645s] [ext: omni.kit.pipapi-0.0.0] startup
[1.649s] [ext: omni.kit.pip_archive-0.0.0] startup
[1.656s] [ext: omni.kit.loop-isaac-1.0.0] startup
[1.656s] [ext: omni.kit.async_engine-0.0.0] startup
[1.658s] [ext: omni.kit.test-0.0.0] startup
[1.664s] [ext: omni.usd.config-1.0.0] startup
[1.669s] [ext: omni.usd.libs-1.0.0] startup
[1.680s] [ext: omni.usd.config-1.0.0] startup
[1.686s] [ext: omni.usd.libs-1.0.0] startup
[1.832s] [ext: omni.isaac.core_archive-2.0.1] startup
[1.840s] [ext: omni.pip.torch-1_13_0-0.1.4] startup
[1.842s] [ext: omni.isaac.ml_archive-1.1.0] startup
[1.843s] [ext: omni.client-0.1.1] startup
[1.843s] [ext: omni.isaac.core_archive-2.0.1] startup
[1.852s] [ext: omni.appwindow-1.0.1] startup
[1.850s] [ext: omni.pip.torch-1_13_0-0.1.4] startup
[1.854s] [ext: omni.kit.renderer.core-0.0.0] startup
[1.853s] [ext: omni.isaac.ml_archive-1.1.0] startup
[1.854s] [ext: omni.client-0.1.1] startup
[1.858s] [ext: omni.kit.renderer.capture-0.0.0] startup
[1.860s] [ext: omni.kit.renderer.imgui-0.0.0] startup
[1.862s] [ext: omni.appwindow-1.0.1] startup
[1.865s] [ext: omni.kit.renderer.core-0.0.0] startup
[1.874s] [ext: omni.kit.renderer.capture-0.0.0] startup
[1.877s] [ext: omni.kit.renderer.imgui-0.0.0] startup
[1.912s] [ext: carb.audio-0.1.0] startup
[1.929s] [ext: carb.audio-0.1.0] startup
[1.932s] [ext: omni.ui-2.12.23] startup
[1.939s] [ext: omni.ui-2.12.23] startup
[1.942s] [ext: omni.uiaudio-1.0.0] startup
[1.944s] [ext: omni.kit.mainwindow-1.0.0] startup
[1.946s] [ext: omni.kit.uiapp-0.0.0] startup
[1.946s] [ext: omni.usd.schema.physics-1.0.0] startup
[1.949s] [ext: omni.uiaudio-1.0.0] startup
[1.951s] [ext: omni.kit.mainwindow-1.0.0] startup
[1.953s] [ext: omni.kit.uiapp-0.0.0] startup
[1.953s] [ext: omni.usd.schema.physics-1.0.0] startup
[1.991s] [ext: omni.usd.schema.audio-0.0.0] startup
[1.999s] [ext: omni.usd.schema.audio-0.0.0] startup
[2.001s] [ext: omni.usd.schema.semantics-0.0.0] startup
[2.007s] [ext: omni.usd.schema.semantics-0.0.0] startup
[2.012s] [ext: omni.usd.schema.omniscripting-1.0.0] startup
[2.020s] [ext: omni.usd.schema.omniscripting-1.0.0] startup
[2.021s] [ext: omni.usd.schema.geospatial-0.0.0] startup
[2.029s] [ext: omni.usd.schema.geospatial-0.0.0] startup
[2.032s] [ext: omni.usd.schema.physx-0.0.0] startup
[2.042s] [ext: omni.usd.schema.physx-0.0.0] startup
[2.061s] [ext: omni.usd.schema.anim-0.0.0] startup
[2.074s] [ext: omni.usd.schema.anim-0.0.0] startup
[2.142s] [ext: omni.usd.schema.omnigraph-1.0.0] startup
[2.151s] [ext: omni.gpucompute.plugins-0.0.0] startup
[2.152s] [ext: omni.hydra.scene_delegate-0.3.0] startup
[2.155s] [ext: omni.usd.schema.omnigraph-1.0.0] startup
[2.159s] [ext: omni.hydra.usdrt_delegate-4.1.1] startup
[2.165s] [ext: omni.gpucompute.plugins-0.0.0] startup
[2.165s] [ext: omni.hydra.scene_delegate-0.3.0] startup
[2.172s] [ext: omni.hydra.usdrt_delegate-4.1.1] startup
[2.175s] [ext: omni.usdphysics-104.1.6-5.1] startup
[2.178s] [ext: omni.kit.window.popup_dialog-2.0.15] startup
[2.185s] [ext: omni.kit.actions.core-1.0.0] startup
[2.187s] [ext: omni.kit.audiodeviceenum-1.0.0] startup
[2.189s] [ext: omni.usdphysics-104.1.6-5.1] startup
[2.189s] [ext: omni.kit.widget.nucleus_connector-1.0.2] startup
[2.192s] [ext: omni.kit.window.popup_dialog-2.0.15] startup
[2.192s] [ext: omni.kit.commands-1.4.5] startup
[2.197s] [ext: omni.kvdb-0.0.0] startup
[2.199s] [ext: omni.kit.actions.core-1.0.0] startup
[2.202s] [ext: omni.kit.audiodeviceenum-1.0.0] startup
[2.200s] [ext: omni.convexdecomposition-104.1.6-5.1] startup
[2.204s] [ext: omni.kit.widget.nucleus_connector-1.0.2] startup
[2.204s] [ext: omni.usd.core-1.0.0] startup
[2.207s] [ext: omni.kit.commands-1.4.5] startup
[2.206s] [ext: omni.timeline-1.0.5] startup
[2.209s] [ext: omni.kit.numpy.common-0.1.0] startup
[2.212s] [ext: omni.kvdb-0.0.0] startup
[2.211s] [ext: omni.usd-1.6.23] startup
[2.215s] [ext: omni.convexdecomposition-104.1.6-5.1] startup
[2.219s] [ext: omni.usd.core-1.0.0] startup
[2.221s] [ext: omni.timeline-1.0.5] startup
[2.224s] [ext: omni.kit.numpy.common-0.1.0] startup
[2.226s] [ext: omni.usd-1.6.23] startup
[2.256s] [ext: omni.localcache-0.0.0] startup
[2.259s] [ext: omni.isaac.version-1.0.0] startup
[2.260s] [ext: omni.physx-104.1.6-5.1] startup
[2.271s] [ext: omni.localcache-0.0.0] startup
[2.274s] [ext: omni.isaac.version-1.0.0] startup
[2.275s] [ext: omni.physx-104.1.6-5.1] startup
2023-05-05 06:13:34 [2,282ms] [Warning] [omni.kvdb.plugin] wasn't able to load the meta database, trying to repair it ...
[2.295s] [ext: omni.kit.widget.prompt-1.0.4] startup
[2.296s] [ext: omni.isaac.dynamic_control-1.2.2] startup
[2.302s] [ext: omni.kit.menu.utils-1.4.7] startup
[2.373s] [ext: omni.kit.widget.path_field-2.0.4] startup
[2.374s] [ext: omni.kit.notification_manager-1.0.4] startup
[2.376s] [ext: omni.kit.search_core-1.0.1] startup
[2.377s] [ext: omni.kit.widget.browser_bar-2.0.4] startup
[2.378s] [ext: omni.kit.widget.versioning-1.3.8] startup
[2.379s] [ext: omni.kit.widget.filebrowser-2.3.8] startup
[2.394s] [ext: omni.kit.collaboration.channel_manager-1.0.9] startup
[2.397s] [ext: omni.iray.libs-0.0.0] startup
[2.400s] [ext: omni.kit.window.filepicker-2.7.8] startup
[2.450s] [ext: omni.kit.usd.layers-2.0.10] startup
[2.458s] [ext: omni.mdl.neuraylib-0.1.0] startup
[2.461s] [ext: omni.kit.window.file_importer-1.0.8] startup
[2.462s] [ext: omni.volume-0.1.0] startup
[2.465s] [ext: omni.mdl-0.1.0] startup
[2.482s] [ext: omni.kit.clipboard-1.0.0] startup
[2.483s] [ext: omni.hydra.rtx-0.1.0] startup
[2.489s] [ext: omni.kit.material.library-1.3.21] startup
[2.494s] [ext: omni.activity.core-1.0.1] startup
[2.500s] [ext: omni.kit.hydra_texture-1.0.11] startup
[2.506s] [ext: omni.kit.menu.create-1.0.8] startup
[2.507s] [ext: omni.kit.viewport.registry-104.0.2] startup
[2.510s] [ext: omni.kit.widget.viewport-104.1.15] startup
[2.511s] [ext: omni.kit.context_menu-1.5.10] startup
[2.515s] [ext: omni.kit.window.file_exporter-1.0.10] startup
[2.516s] [ext: omni.ui.scene-1.5.17] startup
[2.522s] [ext: omni.kit.window.property-1.8.2] startup
[2.523s] [ext: omni.kit.widget.stage-2.7.18] startup
[2.526s] [ext: omni.kit.viewport.window-104.1.18] startup
2023-05-05 06:13:35 [2,520ms] [Warning] [omni.kit.ui] Failed to set value for menu with path: Window/Viewport/Viewport Next 1. It doesn't exist.
[2.553s] [ext: omni.kit.property.usd-3.18.12] startup
[2.561s] [ext: omni.kit.viewport.utility-1.0.12] startup
[2.561s] [ext: omni.kit.widget.graph-1.5.3] startup
[2.592s] [ext: omni.inspect-1.0.1] startup
[2.595s] [ext: omni.kit.widget.text_editor-1.0.2] startup
[2.596s] [ext: omni.ui_query-1.1.1] startup
[2.599s] [ext: omni.kit.window.extensions-1.1.1] startup
[2.602s] [ext: omni.graph.core-2.64.7] startup
[2.605s] [ext: omni.kit.primitive.mesh-1.0.6] startup
[2.609s] [ext: omni.kit.ui_test-1.2.9] startup
[2.611s] [ext: omni.kit.widget.settings-1.0.1] startup
[2.612s] [ext: omni.graph.tools-1.17.0] startup
[2.626s] [ext: omni.kit.usd_undo-0.1.1] startup
[2.628s] [ext: omni.kit.stage_templates-1.1.11] startup
[2.630s] [ext: omni.kit.widget.searchfield-1.0.8] startup
[2.631s] [ext: omni.kit.widget.zoombar-1.0.4] startup
[2.631s] [ext: omni.kit.window.preferences-1.3.7] startup
[2.672s] [ext: omni.graph-1.50.1] startup
[2.785s] [ext: omni.debugdraw-0.1.1] startup
[2.791s] [ext: omni.kit.browser.core-2.2.1] startup
[2.796s] [ext: omni.graph.ui-1.24.1] startup
[2.825s] [ext: omni.kit.window.file-1.3.30] startup
[2.827s] [ext: omni.kit.browser.folder.core-1.7.1] startup
[2.829s] [ext: omni.graph.nodes-1.48.3] startup
[2.840s] [ext: omni.graph.action-1.31.0] startup
[2.847s] [ext: omni.kit.browser.sample-1.2.4] startup
[2.849s] [ext: omni.syntheticdata-0.2.4] startup
2023-05-05 06:13:35 [2,852ms] [Warning] [omni.syntheticdata.scripts.extension] SyntheticData extension needs at least a stageFrameHistoryCount of 3
[2.863s] [ext: omni.physics.tensors-0.1.0] startup
[2.870s] [ext: omni.warp-0.6.1] startup
Warp 0.6.1 initialized:
CUDA Toolkit: 11.5, Driver: 12.1
Devices:
"cpu" | x86_64
"cuda:0" | NVIDIA GeForce RTX 4090 (sm_89)
"cuda:1" | NVIDIA GeForce RTX 4090 (sm_89)
Kernel cache: /home/bizon/.cache/warp/0.6.1
2023-05-05 06:13:35 [2,874ms] [Warning] [omni.kvdb.plugin] repair failed
[2.881s] [ext: omni.kit.widget.prompt-1.0.4] startup
[2.882s] [ext: omni.isaac.dynamic_control-1.2.2] startup
[2.889s] [ext: omni.kit.menu.utils-1.4.7] startup
[2.960s] [ext: omni.kit.widget.path_field-2.0.4] startup
[2.961s] [ext: omni.kit.notification_manager-1.0.4] startup
[2.963s] [ext: omni.kit.search_core-1.0.1] startup
[2.965s] [ext: omni.kit.widget.browser_bar-2.0.4] startup
[2.965s] [ext: omni.kit.widget.versioning-1.3.8] startup
[2.967s] [ext: omni.kit.widget.filebrowser-2.3.8] startup
[2.981s] [ext: omni.kit.collaboration.channel_manager-1.0.9] startup
[2.984s] [ext: omni.iray.libs-0.0.0] startup
[2.986s] [ext: omni.kit.window.filepicker-2.7.8] startup
[3.036s] [ext: omni.kit.usd.layers-2.0.10] startup
[3.043s] [ext: omni.mdl.neuraylib-0.1.0] startup
[3.044s] [ext: omni.graph.scriptnode-0.9.3] startup
[3.047s] [ext: omni.kit.window.file_importer-1.0.8] startup
[3.047s] [ext: omni.volume-0.1.0] startup
[3.046s] [ext: omni.command.usd-1.0.1] startup
[3.050s] [ext: omni.mdl-0.1.0] startup
[3.049s] [ext: omni.physx.tensors-0.1.0] startup
[3.055s] [ext: omni.replicator.core-1.6.4] startup
[3.066s] [ext: omni.kit.clipboard-1.0.0] startup
[3.067s] [ext: omni.hydra.rtx-0.1.0] startup
[3.073s] [ext: omni.kit.material.library-1.3.21] startup
[3.078s] [ext: omni.activity.core-1.0.1] startup
2023-05-05 06:13:35 [3,069ms] [Warning] [omni.replicator.core.scripts.annotators] Annotator PostProcessDispatch is already registered, overwriting annotator template
[3.084s] [ext: omni.kit.hydra_texture-1.0.11] startup
[3.090s] [ext: omni.kit.menu.create-1.0.8] startup
[3.091s] [ext: omni.kit.viewport.registry-104.0.2] startup
[3.093s] [ext: omni.kit.widget.viewport-104.1.15] startup
[3.095s] [ext: omni.kit.context_menu-1.5.10] startup
[3.099s] [ext: omni.kit.window.file_exporter-1.0.10] startup
[3.100s] [ext: omni.ui.scene-1.5.17] startup
[3.105s] [ext: omni.kit.window.property-1.8.2] startup
[3.107s] [ext: omni.kit.widget.stage-2.7.18] startup
[3.110s] [ext: omni.kit.viewport.window-104.1.18] startup
2023-05-05 06:13:35 [3,104ms] [Warning] [omni.kit.ui] Failed to set value for menu with path: Window/Viewport/Viewport Next 1. It doesn't exist.
[3.136s] [ext: omni.kit.property.usd-3.18.12] startup
[3.141s] [ext: omni.isaac.core-1.42.0] startup
[3.144s] [ext: omni.kit.viewport.utility-1.0.12] startup
[3.145s] [ext: omni.kit.widget.graph-1.5.3] startup
[3.174s] [ext: omni.inspect-1.0.1] startup
[3.177s] [ext: omni.kit.widget.text_editor-1.0.2] startup
[3.178s] [ext: omni.ui_query-1.1.1] startup
[3.181s] [ext: omni.kit.window.extensions-1.1.1] startup
[3.184s] [ext: omni.graph.core-2.64.7] startup
[3.188s] [ext: omni.kit.primitive.mesh-1.0.6] startup
[3.191s] [ext: omni.kit.ui_test-1.2.9] startup
[3.193s] [ext: omni.kit.widget.settings-1.0.1] startup
[3.194s] [ext: omni.graph.tools-1.17.0] startup
[3.208s] [ext: omni.kit.usd_undo-0.1.1] startup
[3.209s] [ext: omni.kit.stage_templates-1.1.11] startup
[3.211s] [ext: omni.kit.widget.searchfield-1.0.8] startup
[3.212s] [ext: omni.kit.widget.zoombar-1.0.4] startup
[3.213s] [ext: omni.kit.window.preferences-1.3.7] startup
[3.214s] [ext: omni.graph.bundle.action-1.3.0] startup
[3.214s] [ext: omni.replicator.isaac-1.7.3] startup
[3.224s] [ext: omni.kit.window.title-1.1.2] startup
[3.225s] [ext: omni.isaac.kit-1.2.1] startup
[3.225s] [ext: omni.isaac.sim.python.gym.headless-2022.2.0] startup
[3.226s] Simulation App Starting
[3.253s] [ext: omni.graph-1.50.1] startup
[3.368s] [ext: omni.debugdraw-0.1.1] startup
[3.374s] [ext: omni.kit.browser.core-2.2.1] startup
[3.379s] [ext: omni.graph.ui-1.24.1] startup
[3.408s] [ext: omni.kit.window.file-1.3.30] startup
[3.410s] [ext: omni.kit.browser.folder.core-1.7.1] startup
[3.413s] [ext: omni.graph.nodes-1.48.3] startup
[3.423s] [ext: omni.graph.action-1.31.0] startup
[3.430s] [ext: omni.kit.browser.sample-1.2.4] startup
[3.432s] [ext: omni.syntheticdata-0.2.4] startup
2023-05-05 06:13:36 [3,435ms] [Warning] [omni.syntheticdata.scripts.extension] SyntheticData extension needs at least a stageFrameHistoryCount of 3
[3.446s] [ext: omni.physics.tensors-0.1.0] startup
[3.453s] [ext: omni.warp-0.6.1] startup
Warp 0.6.1 initialized:
CUDA Toolkit: 11.5, Driver: 12.1
Devices:
"cpu" | x86_64
"cuda:0" | NVIDIA GeForce RTX 4090 (sm_89)
"cuda:1" | NVIDIA GeForce RTX 4090 (sm_89)
Kernel cache: /home/bizon/.cache/warp/0.6.1
[3.623s] [ext: omni.graph.scriptnode-0.9.3] startup
[3.625s] [ext: omni.command.usd-1.0.1] startup
[3.628s] [ext: omni.physx.tensors-0.1.0] startup
[3.634s] [ext: omni.replicator.core-1.6.4] startup
2023-05-05 06:13:36 [3,648ms] [Warning] [omni.replicator.core.scripts.annotators] Annotator PostProcessDispatch is already registered, overwriting annotator template
[3.717s] [ext: omni.isaac.core-1.42.0] startup
[3.790s] [ext: omni.graph.bundle.action-1.3.0] startup
[3.790s] [ext: omni.replicator.isaac-1.7.3] startup
[3.800s] [ext: omni.kit.window.title-1.1.2] startup
[3.801s] [ext: omni.isaac.kit-1.2.1] startup
[3.801s] [ext: omni.isaac.sim.python.gym.headless-2022.2.0] startup
[3.801s] Simulation App Starting
[7.783s] app ready
[7.792s] app ready
[7.996s] Simulation App Startup Complete
task:
name: Ant
physics_engine: physx
env:
numEnvs: 4096
envSpacing: 5
episodeLength: 1000
enableDebugVis: False
clipActions: 1.0
powerScale: 0.5
controlFrequencyInv: 2
headingWeight: 0.5
upWeight: 0.1
actionsCost: 0.005
energyCost: 0.05
dofVelocityScale: 0.2
angularVelocityScale: 1.0
contactForceScale: 0.1
jointsAtLimitCost: 0.1
deathCost: -2.0
terminationHeight: 0.31
alive_reward_scale: 0.5
sim:
dt: 0.0083
use_gpu_pipeline: True
gravity: [0.0, 0.0, -9.81]
add_ground_plane: True
add_distant_light: True
use_flatcache: True
enable_scene_query_support: False
disable_contact_processing: False
enable_cameras: False
default_physics_material:
static_friction: 1.0
dynamic_friction: 1.0
restitution: 0.0
physx:
worker_thread_count: 4
solver_type: 1
use_gpu: True
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
contact_offset: 0.02
rest_offset: 0.0
bounce_threshold_velocity: 0.2
friction_offset_threshold: 0.04
friction_correlation_distance: 0.025
enable_sleeping: True
enable_stabilization: True
max_depenetration_velocity: 10.0
gpu_max_rigid_contact_count: 524288
gpu_max_rigid_patch_count: 81920
gpu_found_lost_pairs_capacity: 8192
gpu_found_lost_aggregate_pairs_capacity: 262144
gpu_total_aggregate_pairs_capacity: 8192
gpu_max_soft_body_contacts: 1048576
gpu_max_particle_contacts: 1048576
gpu_heap_capacity: 67108864
gpu_temp_buffer_capacity: 16777216
gpu_max_num_partitions: 8
Ant:
override_usd_defaults: False
enable_self_collisions: False
enable_gyroscopic_forces: True
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
sleep_threshold: 0.005
stabilization_threshold: 0.001
density: -1
max_depenetration_velocity: 10.0
train:
params:
seed: 42
algo:
name: a2c_continuous
model:
name: continuous_a2c_logstd
network:
name: actor_critic
separate: False
space:
continuous:
mu_activation: None
sigma_activation: None
mu_init:
name: default
sigma_init:
name: const_initializer
val: 0
fixed_sigma: True
mlp:
units: [256, 128, 64]
activation: elu
d2rl: False
initializer:
name: default
regularizer:
name: None
load_checkpoint: False
load_path:
config:
name: Ant
full_experiment_name: Ant
env_name: rlgpu
device: cuda:1
device_name: cuda:1
multi_gpu: True
ppo: True
mixed_precision: True
normalize_input: True
normalize_value: True
value_bootstrap: True
num_actors: 4096
reward_shaper:
scale_value: 0.01
normalize_advantage: True
gamma: 0.99
tau: 0.95
learning_rate: 0.0003
lr_schedule: adaptive
schedule_type: legacy
kl_threshold: 0.008
score_to_win: 20000
max_epochs: 500
save_best_after: 100
save_frequency: 50
grad_norm: 1.0
entropy_coef: 0.0
truncate_grads: True
e_clip: 0.2
horizon_length: 16
minibatch_size: 32768
mini_epochs: 4
critic_coef: 2
clip_value: True
seq_len: 4
bounds_loss_coef: 0.0001
task_name: Ant
experiment:
num_envs:
seed: 42
torch_deterministic: False
max_iterations:
physics_engine: physx
pipeline: gpu
sim_device: gpu
device_id: 1
rl_device: cuda:1
multi_gpu: True
num_threads: 4
solver_type: 1
test: False
checkpoint:
headless: True
enable_livestream: False
mt_timeout: 30
wandb_activate: False
wandb_group:
wandb_name: Ant
wandb_entity:
wandb_project: omniisaacgymenvs
Setting seed: 42
Sim params does not have attribute: physx
Sim params does not have attribute: Ant
Pipeline: GPU
Pipeline Device: cuda:1
Sim Device: GPU
[8.015s] Simulation App Startup Complete
task:
name: Ant
physics_engine: physx
env:
numEnvs: 4096
envSpacing: 5
episodeLength: 1000
enableDebugVis: False
clipActions: 1.0
powerScale: 0.5
controlFrequencyInv: 2
headingWeight: 0.5
upWeight: 0.1
actionsCost: 0.005
energyCost: 0.05
dofVelocityScale: 0.2
angularVelocityScale: 1.0
contactForceScale: 0.1
jointsAtLimitCost: 0.1
deathCost: -2.0
terminationHeight: 0.31
alive_reward_scale: 0.5
sim:
dt: 0.0083
use_gpu_pipeline: True
gravity: [0.0, 0.0, -9.81]
add_ground_plane: True
add_distant_light: True
use_flatcache: True
enable_scene_query_support: False
disable_contact_processing: False
enable_cameras: False
default_physics_material:
static_friction: 1.0
dynamic_friction: 1.0
restitution: 0.0
physx:
worker_thread_count: 4
solver_type: 1
use_gpu: True
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
contact_offset: 0.02
rest_offset: 0.0
bounce_threshold_velocity: 0.2
friction_offset_threshold: 0.04
friction_correlation_distance: 0.025
enable_sleeping: True
enable_stabilization: True
max_depenetration_velocity: 10.0
gpu_max_rigid_contact_count: 524288
gpu_max_rigid_patch_count: 81920
gpu_found_lost_pairs_capacity: 8192
gpu_found_lost_aggregate_pairs_capacity: 262144
gpu_total_aggregate_pairs_capacity: 8192
gpu_max_soft_body_contacts: 1048576
gpu_max_particle_contacts: 1048576
gpu_heap_capacity: 67108864
gpu_temp_buffer_capacity: 16777216
gpu_max_num_partitions: 8
Ant:
override_usd_defaults: False
enable_self_collisions: False
enable_gyroscopic_forces: True
solver_position_iteration_count: 4
solver_velocity_iteration_count: 0
sleep_threshold: 0.005
stabilization_threshold: 0.001
density: -1
max_depenetration_velocity: 10.0
train:
params:
seed: 42
algo:
name: a2c_continuous
model:
name: continuous_a2c_logstd
network:
name: actor_critic
separate: False
space:
continuous:
mu_activation: None
sigma_activation: None
mu_init:
name: default
sigma_init:
name: const_initializer
val: 0
fixed_sigma: True
mlp:
units: [256, 128, 64]
activation: elu
d2rl: False
initializer:
name: default
regularizer:
name: None
load_checkpoint: False
load_path:
config:
name: Ant
full_experiment_name: Ant
env_name: rlgpu
device: cuda:0
device_name: cuda:0
multi_gpu: True
ppo: True
mixed_precision: True
normalize_input: True
normalize_value: True
value_bootstrap: True
num_actors: 4096
reward_shaper:
scale_value: 0.01
normalize_advantage: True
gamma: 0.99
tau: 0.95
learning_rate: 0.0003
lr_schedule: adaptive
schedule_type: legacy
kl_threshold: 0.008
score_to_win: 20000
max_epochs: 500
save_best_after: 100
save_frequency: 50
grad_norm: 1.0
entropy_coef: 0.0
truncate_grads: True
e_clip: 0.2
horizon_length: 16
minibatch_size: 32768
mini_epochs: 4
critic_coef: 2
clip_value: True
seq_len: 4
bounds_loss_coef: 0.0001
task_name: Ant
experiment:
num_envs:
seed: 42
torch_deterministic: False
max_iterations:
physics_engine: physx
pipeline: gpu
sim_device: gpu
device_id: 0
rl_device: cuda:0
multi_gpu: True
num_threads: 4
solver_type: 1
test: False
checkpoint:
headless: True
enable_livestream: False
mt_timeout: 30
wandb_activate: False
wandb_group:
wandb_name: Ant
wandb_entity:
wandb_project: omniisaacgymenvs
Setting seed: 42
Sim params does not have attribute: physx
Sim params does not have attribute: Ant
Pipeline: GPU
Pipeline Device: cuda:0
Sim Device: GPU
Task Device: cuda:1
RL device: cuda:1
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/gym/spaces/box.py:84: UserWarning: WARN: Box bound precision lowered by casting to float32
logger.warn(f"Box bound precision lowered by casting to {self.dtype}")
Task Device: cuda:0
RL device: cuda:0
/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/site-packages/gym/spaces/box.py:84: UserWarning: WARN: Box bound precision lowered by casting to float32
logger.warn(f"Box bound precision lowered by casting to {self.dtype}")
[8.827s] [ext: omni.isaac.sim.python.gym.headless-2022.2.0] shutdown
[8.827s] [ext: omni.replicator.isaac-1.7.3] shutdown
[8.827s] [ext: omni.replicator.core-1.6.4] shutdown
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,820ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:41 [8,821ms] [Warning] [carb] [Plugin: omni.replicator.core.plugin] Module /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.replicator.core-1.6.4+104.1.lx64.r.cp37/bin/libomni.replicator.core.plugin.so remained loaded after unload request
[8.833s] [ext: omni.syntheticdata-0.2.4] shutdown
[8.834s] [ext: omni.isaac.core-1.42.0] shutdown
[8.834s] [ext: omni.graph.scriptnode-0.9.3] shutdown
[8.834s] [ext: omni.warp-0.6.1] shutdown
[8.838s] [ext: omni.graph.bundle.action-1.3.0] shutdown
[8.838s] [ext: omni.graph.nodes-1.48.3] shutdown
[8.839s] [ext: omni.graph.action-1.31.0] shutdown
[8.839s] [ext: omni.graph.ui-1.24.1] shutdown
2023-05-05 06:13:41 [8,834ms] [Warning] [omni.kit.ui] Failed to remove menu with path: Window/Visual Scripting/Toolkit. It doesn't exist.
2023-05-05 06:13:41 [8,834ms] [Warning] [omni.kit.menu.utils.scripts.utils] omni.kit.menu.utils remove_menu_items "Create" failed list.remove(x): x not in list
2023-05-05 06:13:41 [8,834ms] [Warning] [omni.kit.ui] Failed to remove menu with path: Window/Visual Scripting/Toolkit. It doesn't exist.
2023-05-05 06:13:41 [8,834ms] [Warning] [omni.kit.ui] Failed to remove menu with path: Window/Visual Scripting/Node Description Editor. It doesn't exist.
[9.077s] [ext: omni.kit.viewport.utility-1.0.12] shutdown
[9.077s] [ext: omni.kit.browser.sample-1.2.4] shutdown
[9.078s] [ext: omni.kit.browser.folder.core-1.7.1] shutdown
[9.079s] [ext: omni.kit.browser.core-2.2.1] shutdown
[9.079s] [ext: omni.kit.viewport.window-104.1.18] shutdown
2023-05-05 06:13:41 [9,094ms] [Warning] [omni.ext._impl._internal] omni.kit.viewport.window-104.1.18 -> <class 'omni.kit.viewport.window.extension.ViewportWindowExtension'>: extension object is still alive, something holds a reference on it. References: ["[0]:type: <class 'frame'>, id: 334733824", "[1]:type: <class 'frame'>, id: 334582032", "[2]:type: <class 'frame'>, id: 111020256"]
[9.448s] [ext: omni.isaac.sim.python.gym.headless-2022.2.0] shutdown
[9.448s] [ext: omni.replicator.isaac-1.7.3] shutdown
[9.448s] [ext: omni.replicator.core-1.6.4] shutdown
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [omni.graph.core.plugin] Could not find category 'Replicator:Annotators' for removal
2023-05-05 06:13:42 [9,441ms] [Warning] [carb] [Plugin: omni.replicator.core.plugin] Module /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.replicator.core-1.6.4+104.1.lx64.r.cp37/bin/libomni.replicator.core.plugin.so remained loaded after unload request
[9.454s] [ext: omni.warp-0.6.1] shutdown
[9.458s] [ext: omni.kit.browser.sample-1.2.4] shutdown
[9.459s] [ext: omni.kit.browser.folder.core-1.7.1] shutdown
[9.459s] [ext: omni.kit.browser.core-2.2.1] shutdown
[9.459s] [ext: omni.syntheticdata-0.2.4] shutdown
[9.460s] [ext: omni.isaac.core-1.42.0] shutdown
[9.460s] [ext: omni.graph.scriptnode-0.9.3] shutdown
[9.460s] [ext: omni.graph.bundle.action-1.3.0] shutdown
[9.460s] [ext: omni.graph.nodes-1.48.3] shutdown
[9.461s] [ext: omni.graph.action-1.31.0] shutdown
[9.461s] [ext: omni.graph.ui-1.24.1] shutdown
2023-05-05 06:13:42 [9,456ms] [Warning] [omni.kit.ui] Failed to remove menu with path: Window/Visual Scripting/Toolkit. It doesn't exist.
2023-05-05 06:13:42 [9,456ms] [Warning] [omni.kit.menu.utils.scripts.utils] omni.kit.menu.utils remove_menu_items "Create" failed list.remove(x): x not in list
2023-05-05 06:13:42 [9,456ms] [Warning] [omni.kit.ui] Failed to remove menu with path: Window/Visual Scripting/Toolkit. It doesn't exist.
2023-05-05 06:13:42 [9,456ms] [Warning] [omni.kit.ui] Failed to remove menu with path: Window/Visual Scripting/Node Description Editor. It doesn't exist.
[9.697s] [ext: omni.kit.viewport.utility-1.0.12] shutdown
[9.697s] [ext: omni.kit.viewport.window-104.1.18] shutdown
2023-05-05 06:13:42 [9,713ms] [Warning] [omni.ext._impl._internal] omni.kit.viewport.window-104.1.18 -> <class 'omni.kit.viewport.window.extension.ViewportWindowExtension'>: extension object is still alive, something holds a reference on it. References: ["[0]:type: <class 'frame'>, id: 343270400", "[1]:type: <class 'frame'>, id: 343118608", "[2]:type: <class 'frame'>, id: 119418496"]
Error executing job with overrides: ['headless=True', 'task=Ant', 'multi_gpu=True']
Traceback (most recent call last):
File "scripts/rlgames_train.py", line 115, in parse_hydra_configs
task = initialize_task(cfg_dict, env)
File "/home/bizon/eric/OmniIsaacGymEnvs/omniisaacgymenvs/utils/task_util.py", line 72, in initialize_task
env.set_task(task=task, sim_params=sim_config.get_physics_params(), backend="torch", init_sim=init_sim)
File "/home/bizon/eric/OmniIsaacGymEnvs/omniisaacgymenvs/envs/vec_env_rlgames.py", line 51, in set_task
super().set_task(task, backend, sim_params, init_sim)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.isaac.gym/omni/isaac/gym/vec_env/vec_env_base.py", line 80, in set_task
self._world.reset()
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/exts/omni.isaac.core/omni/isaac/core/world/world.py", line 285, in reset
task.post_reset()
File "/home/bizon/eric/OmniIsaacGymEnvs/omniisaacgymenvs/tasks/ant.py", line 86, in post_reset
LocomotionTask.post_reset(self)
File "/home/bizon/eric/OmniIsaacGymEnvs/omniisaacgymenvs/tasks/shared/locomotion.py", line 174, in post_reset
self.reset_idx(indices)
File "/home/bizon/eric/OmniIsaacGymEnvs/omniisaacgymenvs/tasks/shared/locomotion.py", line 125, in reset_idx
self.initial_dof_pos[env_ids] + dof_pos, self.dof_limits_lower, self.dof_limits_upper
RuntimeError: indices should be either on cpu or on the same device as the indexed tensor (cuda:0)

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
self.seed = 42
Started to train
Exact experiment name requested from command line: Ant
WARNING:torch.distributed.elastic.multiprocessing.api:Sending process 321900 closing signal SIGTERM
WARNING:torch.distributed.elastic.multiprocessing.api:Unable to shutdown process 321900 via 15, forcefully exitting via 9
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -11) local_rank: 1 (pid: 321901) of binary: /home/bizon/anaconda3/envs/isaac-sim/bin/python
Traceback (most recent call last):
File "/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/runpy.py", line 193, in _run_module_as_main
"main", mod_spec)
File "/home/bizon/anaconda3/envs/isaac-sim/lib/python3.7/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.pip.torch-1_13_0-0.1.4+104.1.lx64/torch-1-13-0/torch/distributed/run.py", line 766, in
main()
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.pip.torch-1_13_0-0.1.4+104.1.lx64/torch-1-13-0/torch/distributed/elastic/multiprocessing/errors/init.py", line 346, in wrapper
return f(*args, **kwargs)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.pip.torch-1_13_0-0.1.4+104.1.lx64/torch-1-13-0/torch/distributed/run.py", line 762, in main
run(args)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.pip.torch-1_13_0-0.1.4+104.1.lx64/torch-1-13-0/torch/distributed/run.py", line 756, in run
)(*cmd_args)
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.pip.torch-1_13_0-0.1.4+104.1.lx64/torch-1-13-0/torch/distributed/launcher/api.py", line 132, in call
return launch_agent(self._config, self._entrypoint, list(args))
File "/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.pip.torch-1_13_0-0.1.4+104.1.lx64/torch-1-13-0/torch/distributed/launcher/api.py", line 248, in launch_agent
failures=result.failures,
torch.distributed.elastic.multiprocessing.errors.ChildFailedError:

scripts/rlgames_train.py FAILED

Failures:
<NO_OTHER_FAILURES>

Root Cause (first observed failure):
[0]:
time : 2023-05-04_20:13:55
host : dl
rank : 1 (local_rank: 1)
exitcode : -11 (pid: 321901)
error_file: <N/A>
traceback : Signal 11 (SIGSEGV) received by PID 321901

crash with unexpected keyword argument 'enable_livestream' when trying to run 'cartpole'

python scripts/rlgames_train.py task=Cartpole

/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/extscache/omni.pip.torch-1_13_0-0.1.4+104.1.lx64/torch-1-13-0/torch/utils/tensorboard/init.py:5: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
tensorboard.version
/home/bizon/anaconda3/envs/issac_sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config': Defaults list is missing _self_. See https://hydra.cc/docs/upgrades/1.0_to_1.1/default_composition_order for more information
warnings.warn(msg, UserWarning)
/home/bizon/anaconda3/envs/issac_sim/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:412: UserWarning: In config: Invalid overriding of hydra/job_logging:
Default list overrides requires 'override' keyword.
See https://hydra.cc/docs/next/upgrades/1.0_to_1.1/defaults_list_override for more information.

deprecation_warning(msg)
Error executing job with overrides: ['task=Cartpole']
Traceback (most recent call last):
File "scripts/rlgames_train.py", line 98, in parse_hydra_configs
env = VecEnvRLGames(headless=headless, sim_device=cfg.device_id, enable_livestream=cfg.enable_livestream, enable_viewport=enable_viewport)
TypeError: init() got an unexpected keyword argument 'enable_livestream'

I am running in a conda enviornment which is why I am using 'python' to launch.
Any ideas? I am completely stuck here.

Feature Request: migrate from gym to gymnasium

Hi, would it be possible for IsaacGymEnvs to be upgraded from gym to gymnasium? Gymnasium is the maintained version of openai gym and is compatible with current RL training libraries (rllib and tianshou have already migrated, and stable-baselines3 will soon).

This repository is currently listed in the gymnasium third party environments but we are cleaning the list up to only include maintained gymnasium-compatible repositories.

For information about upgrading and compatibility, see migration guide and gym compatibility. The main difference is the API has switched to returning truncated and terminated, rather than done, in order to give more information and mitigate edge case issues.

When using Go1 USD model collisions do not work.

If I use the go1.usd instead of Anymal in the Anymal task (I have carefully set up all the joints correctly) It falls right though the ground plane during training. If I use the a1.usd instead it doesn't fall though the ground plane during training but does fall though during testing. Ironically the go1.usd does the opposite. It falls though on training but works fine during testing. What's going on here? How to go about debugging the collision detection.

Only thing I can think of is there is the warning:

[Warning] [omni.physx.plugin] PhysicsUSD: Prim at path /World/envs/env_0/go1/RL_hip/collisions is using obsolete 'customGeometry' attribute. To toggle custom geometry for cylinders, use the physics settings option.

How do you set this setting?

links coordination do not follow robot's movement

Hi all,

I met one problem that the link's coordination keeps static when robot base is moving as the video.
$ PYTHON_PATH scripts/random_policy.py task=Ant num_envs=1
It causes a big problem when adding sensors to the link, for example the IMU sensor will always output lin_acc [0, 0, 9.8] and ang_vel [0, 0, 0] when adding to the torso.

Meanwhile I checked IMU example in Isaac Sim, and the coordinate worked perfectly as the video.

Failed to find the reason after checking codes. I really appreciate it if you can give some hints.

Not possible to run on any GPU other than cuda:0

on line 106 of simulation_context.py

   if self._device is not None and "cuda" in self._device:
        device_id = self._settings.get_as_int("/physics/cudaDevice")
        self._device = f"cuda:{device_id}"

this will always return cuda:0 as far as I can tell. No way to pass in cuda:1 when constructing the context.

I have changed both device_id and rl_device in config.yaml in omniIsaacgymenvs and config.yaml in AnymalTerrain. So cuda: 1 seems to be used everywhere else but here.

conda instructions for installations aren't working

Following the instructions for installing a conda enviorment. Here is what I typed:

(issac_sim) bizon@dl:/eric/OmniIsaacGymEnvs$ source ~/.local/share/ov/pkg/isaac_sim-2022.2.0/setup_conda_env.sh
(issac_sim) bizon@dl:/eric/OmniIsaacGymEnvs$ alias PYTHON_PATH=~/.local/share/ov/pkg/isaac_sim-2022.2.0/python.sh
(issac_sim) bizon@dl:/eric/OmniIsaacGymEnvs$ PYTHON_PATH -m pip install -e .
Warning: running in conda env, please deactivate before executing this script
If conda is desired please source setup_conda_env.sh in your python 3.7 conda env and run python normally
/home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/python.sh: line 41: /home/bizon/.local/share/ov/pkg/isaac_sim-2022.2.0/kit/python/bin/python3: Argument list too long
There was an error running python
(issac_sim) bizon@dl:/eric/OmniIsaacGymEnvs$ ls
docker docs LICENSE.txt omniisaacgymenvs README.md setup.py
(issac_sim) bizon@dl:~/eric/OmniIsaacGymEnvs$ python
Python 3.7.16 (default, Jan 17 2023, 22:20:44)
[GCC 11.2.0] :: Anaconda, Inc. on linux
Type "help", "copyright", "credits" or "license" for more information.

quit()

Something seems to be going wrong in python.sh at the very last line. Any ideas? I am trying to move from issac gym to sim and I am completely stuck..

Can IsaacGym provide a parallel environment class that supports the SMDP or HRL algorithm?

Thanks for the amazing project.

I want to write an environment that supports SMDP or HRL algorithms based on base_task.py.

Take HRL as an example, the RL controller may require each subenvironment in parallel to implement a skill primitive, but the step lengths of the different skill primitives are not the same, which results in all the subenvironments stepping at the same time but ending at different steps.

However, the subenvironment that finishes first continues to step according to the skill primitives provided by RL, regardless of whether the other subenvironments are still executing the previous skill primitives.

Can the current isaacgym environment achieve this asynchronous parallel effect?

Thanks for any suggestion :) @kellyguo11

How to retrieve updated Xform transformation

Hi,

I just wonder if there's good way to get the updated transformation info of Xforms in the viewport? I checked the RL examples (eg. humanoids) and found the transformation info seems to be frozen in the Property (the part within red box in the attached figure) when objects move. xform.GetLocalTransformation() in code also return the same frozen transformation matrix even though it would work for Issac Sim.

Net Contact Forces is always Zero for Anymal Robot

It seems that the contact forces of the Anymal's knees are not correct values, and in the task AnymalTerrain, the reset is not triggered while the knees contact with the terrain.

It is simple to reproduce the issue:

1. Change the defaultJointAngles in AnymalTerrain.yaml to:
   LF_HAA: 0.03 # [rad]
   LH_HAA: 0.03 # [rad]
   RF_HAA: -0.03 # [rad]
   RH_HAA: -0.03 # [rad]

   LF_HFE: 0.4 # [rad]
   LH_HFE: -0.4 # [rad]
   RF_HFE: 0.4 # [rad]
   RH_HFE: -0.4 # [rad]

   LF_KFE: -2.8 # [rad]
   LH_KFE: 2.8 # [rad]
   RF_KFE: -2.8 # [rad]
   RH_KFE: 2.8 # [rad]

2. Run the following command in the terminal:
   python scripts/rlgames_train.py task=AnymalTerrain headless=false num_envs=2 test=True

We will see that the knees of the Anymal robots touches the ground but the env is not terminated, and I found that the returned value of self._anymals._knees.get_net_contact_forces(clone=False) is always 0.

underwater environment

Hi, thank you for the amazing project. Can we create an underwater environment for the robot in the Omniverse Gym? Thank you.

[Error] There was an error when I runned examples

After Installation, I tried to launch rlgames_train.py

~/OmniIsaacGymEnvs/omniisaacgymenvs$ PYTHON_PATH scripts/rlgames_train.py task=Cartpole

But here is error:

Traceback (most recent call last):
File "scripts/rlgames_train.py", line 34, in
from omniisaacgymenvs.utils.config_utils.path_utils import retrieve_checkpoint_path
File "~/OmniIsaacGymEnvs/omniisaacgymenvs/utils/config_utils/path_utils.py", line 32, in
import omni.client
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/extscore/omni.client/omni/client/init.py", line 21, in
from ._omniclient import *
ImportError: ~ /.local/share/ov/pkg/isaac_sim-2022.1.0/kit/extscore/omni.usd.libs/bin/libjs.so: undefined symbol: _ZN32pxrInternal_v0_20__pxrReserved__18Tf_PostErrorHelperERKNS_13TfCallContextENS_16TfDiagnosticTypeEPKcz
There was an error running python

So I tested the simplest code
Print_pi.py

from omni.isaac.examples.hello_world.hello_world_extension import HelloWorldExtension
print('Hi")

Run
$ PYTHON_PATH print_hi.py

ERROR

Traceback (most recent call last):
File "print_hi.py", line 12, in
from omni.isaac.examples.hello_world.hello_world_extension import HelloWorldExtension
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/exts/omni.isaac.examples/omni/isaac/examples/hello_world/init.py", line 10, in
from omni.isaac.examples.hello_world.hello_world import HelloWorld
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/exts/omni.isaac.examples/omni/isaac/examples/hello_world/hello_world.py", line 9, in
from omni.isaac.examples.base_sample import BaseSample
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/exts/omni.isaac.examples/omni/isaac/examples/base_sample/init.py", line 10, in
from omni.isaac.examples.base_sample.base_sample import BaseSample
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/exts/omni.isaac.examples/omni/isaac/examples/base_sample/base_sample.py", line 9, in
from omni.isaac.core import World
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/exts/omni.isaac.core/omni/isaac/core/init.py", line 9, in
from omni.isaac.core.physics_context.physics_context import PhysicsContext
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/exts/omni.isaac.core/omni/isaac/core/physics_context/init.py", line 9, in
from omni.isaac.core.physics_context.physics_context import PhysicsContext
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/exts/omni.isaac.core/omni/isaac/core/physics_context/physics_context.py", line 12, in
from pxr import Usd, UsdGeom, Gf, Sdf, UsdPhysics, PhysxSchema, UsdShade
File "~/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/extscore/omni.usd.libs/pxr/Usd/init.py", line 24, in
from . import bong_usd
ImportError: ~/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/extscore/omni.usd.libs/bin/libjs.so: undefined symbol: _ZN32pxrInternal_v0_20__pxrReserved__18Tf_PostErrorHelperERKNS_13TfCallContextENS_16TfDiagnosticTypeEPKcz
There was an error running python

How can I solve it?

Here's my hardware spec
Ubuntu 20.04
Python 3.8 (I tried Python 3.7 in conda env, but I got the same result)
Intel CPU + RTX3070Ti

Support for humanoidAMP

Hello,

Thanks for providing this amazing repo.
Are there any plans to add/migrate the implementation of the HumanoidAMP task from IsaacGymEnvs repo?
If so, what's the timeline?

Different Agent Performance Between CPU training and GPU training

Hi,

I tried to train an agent performing a custom RL task, but I found that the performances and behaviors of the agent are different, depending on the training device (CPU or GPU).

In general, with all other parameters, as well as task codes the same, CPU trained agent has noticeable better performance that GPU trained agent (red: CPU trained, purple: GPU trained):

And I have defined a success condition, the success rate of CPU trained agent is far better (about 90%) than GPU (about 50%):

I have trained on CPU for many times, with only minor differences (changed some reward scales for example), and the success rates are always high (above 90%); But when I switched to GPU, I trained twice but got 50% success rates both.

Another thing I noticed is that the robots trained using GPU are shaking very obviously, which is not a desired behavior since the success condition requires it to have very low joint velocities, base linear&angular velocities. CPU trained agents have an extremely low velocities near success regions, but I checked the tensorboard, the average joint acceleration of CPU trained is actually larger than GPU (and I wonder why the joint acceleration is optimized more than other important reward terms):

So my questions are:

1. Is GPU simulation very different from CPU simulation?
2. Are there any parameters that I can change to make GPU simulation close to CPU?
3. I used built-in velocity controller to control robot joints, does the controller differ between GPU and CPU?

Thanks in advance!

Lidar collision detection isn't working properly

Hi, we're trying to use this repository to train multiple robots that use Lidar sensors for detecting obstacles. For testing if the sensors work correctly, we added a cube (with the collision API enabled) to the default environment. When running the simulation with the random policy script however, the Lidars don't seem to detect any obstacles. This can be remedied by selecting the cube in the GUI and disabling and enabling the collision API of the cube again, but this workaround only works for objects in the default environment and not for any of the clones.

This is a screenshot of the scene after starting the simulation

This is a screenshot of the scene after disabling and reenabling the collision api of the cube in env_0

Is there any way to configure the scene so the Lidar sensors work properly in every environment without having to use the GUI?

[Error] Memory allocation error when running examples involving hand models

Hi,

I encounter an issue when trying to run examples you provide that involve hand models, i.e. either the Allegro hand of the Shadow one. I tried to run several different commands, but I get the same error every time:

PxgCudaDeviceMemoryAllocator fail to allocate memory 2147483648 bytes!! Result = 2

I tried to run scripts/random_policy.py and scripts/rlgames_train.py, using num_envs=1 (and more, but I assume that 1 is the lighter choice in terms of memory), with test=True (inference mode) as well as test=False (training mode), and with headless=True as well as headless=False. For all these, I get the exact same error mentioned above (always the same number of bytes, i.e. 2 GB), followed by lots of PhysX errors and a Exception: Failed to create simulation view backend exception at the end, and a segfault. The whole output is quite large, but I could paste it here if needed.

I tried running other examples (Anymal, AnymalTerrain, Cartpole, etc.) and they happen to work, even for multiple envs (e.g. Anymal with 50 robots). Bad luck, I'm working with robotic hands...

I did monitor my GPU to see what was happening concerning memory, and the GPU memory usage goes to a bit more than 4GB (for 6GB of available memory) then the program crashes. I assume the problem comes from that, since 2GB is thus more than available.

I don't really know where these 2GB come from, and why this seems to only happen with the two hands. This issue seems related to mine, if that can help.

I don't know if such an issue is expected considering my specs (see below), or if you experienced such problems on your end, but since I'm planning to use Isaac Gym to do RL with the shadow hand, I'd appreciate your opinion on this problem. Feel free to ask for more info, including full terminal output if needed.

Thank you in advance for your help!

---

My environment:
Commit: 1aaf354
Isaac Sim Version: 2022.2.0
OS: Ubuntu 20.04
GPU: RTX 3060
CUDA: 12.0
GPU Driver: 525.85.05

question about prim_paths

Hi,
Thanks for the great jobs on Omniverse Isaac Gym.
I have a question when I transfer policy from isaac gym. I used my own asset so I get a urdf file.
I use isaac sim to import it as a usd file, but there is no prim_paths on isaac sim to save.
So when I want to train by omni isaac gym, it shows wrong. How can I get prim file?

Examples segfault

Hi,

I'm trying to run some of the examples. While the Ant task works fine, other robots lead to a segfault during the initialize_task() function. This includes both the cartpole and franke_cabinet tasks.

I am running this in the latest Isaac docker image nvcr.io/nvidia/isaac-sim:2022.1.1 on an RTX A6000 with 515.65.1 drivers

The error is

[Warning] [omni.isaac.kit.simulation_app] Modules: ['omniisaacgymenvs', 'omniisaacgymenvs.utils', 'omniisaacgymenvs.utils.hydra_cfg', 'omniisaacgymenvs.utils.hydra_cfg.hydra_utils', 'omniisaacgymenvs.utils.hydra_cfg.reformat', 'omniisaacgymenvs.utils.rlgames', 'omniisaacgymenvs.utils.rlgames.rlgames_utils', 'omniisaacgymenvs.utils.config_utils', 'omniisaacgymenvs.utils.config_utils.path_utils', 'omniisaacgymenvs.envs', 'omniisaacgymenvs.envs.vec_env_rlgames', 'omni.isaac.gym', 'omni.isaac.gym.vec_env', 'omni.isaac.gym.vec_env.vec_env_base', 'omni.isaac.gym.vec_env.vec_env_mt'] were loaded before SimulationApp was started and might not be loaded correctly.
[Warning] [omni.isaac.kit.simulation_app] Please check to make sure no extra omniverse or pxr modules are imported before the call to SimulationApp(...)
Starting kit application with the fillowing args:  ['/isaac-sim/exts/omni.isaac.kit/omni/isaac/kit/simulation_app.py', '/isaac-sim/apps/omni.isaac.sim.python.kit', '--/app/tokens/exe-path=/isaac-sim/kit', '--/persistent/app/viewport/displayOptions=3094', '--/rtx/materialDb/syncLoads=True', '--/rtx/hydra/materialSyncLoads=True--/omni.kit.plugin/syncUsdLoads=True', '--/app/renderer/resolution/width=1280', '--/app/renderer/resolution/height=720', '--/app/window/width=1440', '--/app/window/height=900', '--/renderer/multiGpu/enabled=True', '--ext-folder', '/isaac-sim/exts', '--ext-folder', '/isaac-sim/apps', '--/physics/cudaDevice=0', '--portable', '--allow-root']
Passing the following args to the base kit application:  ['--enable', 'omni.kit.livestream.native', '--no-window']
[Warning] [omni.kit.app.plugin] No crash reporter present, dumps uploading isn't available.
[Info] [carb] Logging to file: /isaac-sim/kit/logs/Kit/Isaac-Sim/2022.1/kit_20221209_085428.log
2022-12-09 08:54:28 [1ms] [Warning] [omni.ext.plugin] [ext: omni.drivesim.sensors.nv.lidar] Extensions config 'extension.toml' doesn't exist '/isaac-sim/exts/omni.drivesim.sensors.nv.lidar' or '/isaac-sim/exts/omni.drivesim.sensors.nv.lidar/config'
[0.115s] [ext: omni.stats-0.0.0] startup
[0.155s] [ext: omni.gpu_foundation-0.0.0] startup
2022-12-09 08:54:28 [153ms] [Warning] [carb] FrameworkImpl::setDefaultPlugin(client: omni.gpu_foundation_factory.plugin, desc : [carb::graphics::Graphics v2.5], plugin : carb.graphics-vulkan.plugin) failed. Plugin selection is locked, because the interface was previously acquired by: 
[0.166s] [ext: carb.windowing.plugins-1.0.0] startup
2022-12-09 08:54:28 [159ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:54:28 [159ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
[0.167s] [ext: omni.assets.plugins-0.0.0] startup
[0.168s] [ext: omni.kit.renderer.init-0.0.0] startup

|---------------------------------------------------------------------------------------------|
| Driver Version: 515.65.1      | Graphics API: Vulkan
|=============================================================================================|
| GPU | Name                             | Active | LDA | GPU Memory | Vendor-ID | LUID       |
|     |                                  |        |     |            | Device-ID | UUID       |
|---------------------------------------------------------------------------------------------|
| 0   | NVIDIA RTX A6000                 | Yes: 0 |     | 49386   MB | 10de      | 0          |
|     |                                  |        |     |            | 2230      | 31d81b8f.. |
|=============================================================================================|
| OS: Linux rlsim2, Version: 5.4.0-132-generic
| Processor: AMD Ryzen Threadripper PRO 3975WX 32-Cores      | Cores: Unknown | Logical: 64
|---------------------------------------------------------------------------------------------|
| Total Memory (MB): 128729 | Free Memory: 119711
| Total Page/Swap (MB): 8191 | Free Page/Swap: 8191
|---------------------------------------------------------------------------------------------|
2022-12-09 08:55:02 [34,347ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:02 [34,347ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
2022-12-09 08:55:02 [34,348ms] [Error] [carb.glinterop.plugin] GLInteropContext::init: carb::windowing is not available
2022-12-09 08:55:02 [34,348ms] [Warning] [gpu.foundation.plugin] Realm: no OpenGL interop context.
2022-12-09 08:55:02 [34,439ms] [Warning] [carb.cudainterop.plugin] On Linux only, CUDA and the display driver does not support IOMMU-enabled bare-metal PCIe peer to peer memory copy.
However, CUDA and the display driver does support IOMMU via VM pass through. As a consequence, users on Linux,
when running on a native bare metal system, should disable the IOMMU. The IOMMU should be enabled and the VFIO driver
be used as a PCIe pass through for virtual machines.
[34.456s] [ext: omni.kit.pipapi-0.0.0] startup
[34.477s] [ext: omni.kit.pip_archive-0.0.0] startup
[34.489s] [ext: omni.isaac.core_archive-1.2.0] startup
[34.533s] [ext: omni.usd.config-1.0.0] startup
[34.539s] [ext: omni.usd.libs-1.0.0] startup
[34.645s] [ext: omni.kit.pip_torch-1_11_0-0.1.3] startup
[34.713s] [ext: omni.isaac.ml_archive-1.1.0] startup
[34.720s] [ext: omni.kit.loop-isaac-0.1.0] startup
[34.721s] [ext: omni.kit.async_engine-0.0.0] startup
[34.725s] [ext: omni.appwindow-1.0.0] startup
2022-12-09 08:55:02 [34,718ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:02 [34,718ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
[34.731s] [ext: omni.client-0.1.0] startup
[34.744s] [ext: omni.kit.test-0.0.0] startup
[34.747s] [ext: omni.kit.renderer.core-0.0.0] startup
2022-12-09 08:55:02 [34,741ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:02 [34,741ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
2022-12-09 08:55:02 [34,742ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:02 [34,742ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
2022-12-09 08:55:02 [34,742ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:02 [34,742ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
2022-12-09 08:55:02 [34,743ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:02 [34,743ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
[34.841s] [ext: omni.ui-2.10.3] startup
[34.864s] [ext: carb.audio-0.1.0] startup
[34.868s] [ext: omni.kit.mainwindow-0.0.0] startup
[34.872s] [ext: omni.uiaudio-1.0.0] startup
[34.875s] [ext: omni.kit.uiapp-0.0.0] startup
[34.875s] [ext: omni.usd.schema.physics-1.0.0] startup
[34.935s] [ext: omni.usd.schema.isaac-0.2.0] startup
[35.029s] [ext: omni.usd.schema.omnigraph-1.0.0] startup
[35.049s] [ext: omni.usd.schema.physx-0.0.0] startup
[35.150s] [ext: omni.usd.schema.anim-0.0.0] startup
[35.264s] [ext: omni.usd.schema.audio-0.0.0] startup
[35.281s] [ext: omni.usd.schema.semantics-0.0.0] startup
[35.299s] [ext: omni.usd.schema.forcefield-0.0.0] startup
[35.318s] [ext: omni.timeline-1.0.2] startup
[35.323s] [ext: omni.hydra.scene_delegate-0.2.0] startup
[35.329s] [ext: omni.kit.commands-1.2.2] startup
[35.342s] [ext: omni.kit.audiodeviceenum-1.0.0] startup
[35.344s] [ext: omni.usd-1.5.3] startup
[35.396s] [ext: omni.kit.asset_converter-1.2.31] startup
[35.418s] [ext: omni.kit.widget.path_field-2.0.3] startup
[35.427s] [ext: omni.kit.search_core-1.0.2] startup
[35.430s] [ext: omni.kit.window.popup_dialog-2.0.8] startup
[35.440s] [ext: omni.kit.widget.browser_bar-2.0.3] startup
[35.443s] [ext: omni.kit.widget.filebrowser-2.2.27] startup
[35.452s] [ext: omni.mdl.neuraylib-0.1.0] startup
[35.460s] [ext: omni.kit.menu.utils-1.2.11] startup
[35.476s] [ext: omni.kit.notification_manager-1.0.5] startup
[35.481s] [ext: omni.kit.widget.versioning-1.3.8] startup
[35.485s] [ext: omni.kit.menu.create-1.0.2] startup
[35.486s] [ext: omni.kit.window.filepicker-2.4.30] startup
OmniAssetFileFormat
[35.573s] [ext: omni.mdl-0.1.0] startup
[35.601s] [ext: omni.kit.stage_templates-1.1.2] startup
[35.606s] [ext: omni.kit.window.file_exporter-1.0.4] startup
[35.608s] [ext: omni.kit.material.library-1.3.10] startup
[35.614s] [ext: omni.kit.window.file_importer-1.0.4] startup
[35.615s] [ext: omni.kit.window.drop_support-1.0.0] startup
[35.617s] [ext: omni.kit.window.file-1.3.16] startup
[35.620s] [ext: omni.kit.context_menu-1.3.9] startup
[35.627s] [ext: omni.kit.window.property-1.6.3] startup
[35.629s] [ext: omni.kit.window.content_browser-2.4.28] startup
[35.652s] [ext: omni.kit.widget.stage-2.6.15] startup
[35.659s] [ext: omni.renderer-rtx-0.0.0] startup
[35.659s] [ext: omni.kit.property.usd-3.14.9] startup
[35.731s] [ext: omni.kit.widget.settings-1.0.0] startup
[35.736s] [ext: omni.kit.widget.prompt-1.0.1] startup
[35.737s] [ext: omni.kit.widget.graph-1.4.3] startup
[35.754s] [ext: omni.kit.window.preferences-1.2.1] startup
[35.819s] [ext: omni.hydra.engine.stats-1.0.0] startup
[35.829s] [ext: omni.hydra.rtx-0.1.0] startup
[35.837s] [ext: omni.kit.viewport.legacy_gizmos-1.0.0] startup
[35.839s] [ext: omni.graph.core-2.29.1] startup
[35.842s] [ext: omni.graph.tools-1.4.0] startup
[35.952s] [ext: omni.kit.window.viewport-0.0.0] startup
[39.336s] [ext: omni.graph-1.23.0] startup
[39.439s] [ext: omni.debugdraw-0.1.0] startup
[39.448s] [ext: omni.ui_query-1.1.1] startup
[39.454s] [ext: omni.graph.ui-1.6.1] startup
[39.488s] [ext: omni.kit.ui_test-1.2.2] startup
[39.493s] [ext: omni.kvdb-0.0.0] startup
[39.500s] [ext: omni.graph.action-1.18.0] startup
[39.517s] [ext: omni.kit.widget.searchfield-1.0.6] startup
[39.520s] [ext: omni.convexdecomposition-1.4.15] startup
[39.527s] [ext: omni.localcache-0.0.0] startup
[39.535s] [ext: omni.usdphysics-1.4.15] startup
[39.537s] [ext: omni.graph.scriptnode-0.5.0] startup
[39.541s] [ext: omni.physx-1.4.15-5.1] startup
[39.575s] [ext: omni.kit.numpy.common-0.1.0] startup
[39.584s] [ext: omni.physics.tensors-0.1.0] startup
[39.596s] [ext: omni.graph.nodes-1.26.0] startup
[39.615s] [ext: omni.isaac.dynamic_control-1.1.0] startup
[39.625s] [ext: omni.isaac.version-1.0.0] startup
[39.627s] [ext: omni.physx.tensors-0.1.0] startup
[39.631s] [ext: omni.syntheticdata-0.2.1] startup
[39.692s] [ext: omni.isaac.core-1.24.3] startup
[39.937s] [ext: omni.isaac.synthetic_utils-0.3.5] startup
[39.942s] [ext: omni.kit.window.script_editor-1.6.2] startup
[39.949s] [ext: omni.kit.usd_undo-0.1.0] startup
[39.954s] [ext: semantics.schema.editor-0.2.3] startup
[39.963s] [ext: omni.physx.commands-1.4.15-5.1] startup
[39.974s] [ext: omni.kit.renderer.capture-0.0.0] startup
[39.982s] [ext: omni.physx.ui-1.4.15-5.1] startup
[40.078s] [ext: omni.warp-0.2.2] startup
Warp initialized:
   Version: 0.2.2
   CUDA device: NVIDIA RTX A6000
   Kernel cache: /root/.cache/warp/0.2.2
[40.508s] [ext: omni.kit.property.material-1.8.5] startup
[40.512s] [ext: omni.physx.demos-1.4.15-5.1] startup
[40.520s] [ext: omni.kit.window.toolbar-1.2.4] startup
[40.528s] [ext: omni.kit.property.physx-0.1.0] startup
2022-12-09 08:55:08 [40,674ms] [Warning] [omni.physx.plugin] Deprecated: getSimulationEventStream is deprecated, please use getSimulationEventStreamV2
[40.685s] [ext: omni.ui.scene-1.5.0] startup
[40.697s] [ext: omni.physx.tests-1.4.15-5.1] startup
[41.210s] [ext: omni.physx.vehicle-1.4.15-5.1] startup
[41.233s] [ext: omni.kit.manipulator.viewport-1.0.6] startup
[41.238s] [ext: omni.physx.camera-1.4.15-5.1] startup
[41.250s] [ext: omni.physx.cct-1.4.15-5.1] startup
[41.441s] [ext: omni.command.usd-1.0.1] startup
[41.450s] [ext: omni.kit.primitive.mesh-1.0.0] startup
[41.459s] [ext: omni.graph.visualization.nodes-1.1.1] startup
[41.465s] [ext: omni.physx.bundle-1.4.15-5.1] startup
[41.465s] [ext: omni.kit.widget.stage_icons-1.0.2] startup
[41.468s] [ext: omni.replicator.core-1.4.3] startup
[41.597s] [ext: omni.kit.widget.zoombar-1.0.3] startup
[41.599s] [ext: omni.rtx.window.settings-0.6.1] startup
[41.611s] [ext: omni.kit.window.stage-2.3.7] startup
[41.616s] [ext: omni.replicator.composer-1.1.3] startup
[41.625s] [ext: omni.kit.browser.core-2.0.12] startup
[41.635s] [ext: omni.rtx.settings.core-0.5.5] startup
[41.641s] [ext: omni.kit.window.extensions-1.1.0] startup
[41.650s] [ext: omni.kit.graph.usd.commands-1.1.0] startup
[41.656s] [ext: omni.kit.browser.folder.core-1.1.13] startup
[41.660s] [ext: omni.isaac.debug_draw-0.1.2] startup
[41.667s] [ext: omni.isaac.ui-0.2.1] startup
[41.672s] [ext: omni.kit.graph.delegate.default-1.0.15] startup
[41.673s] [ext: omni.isaac.motion_planning-0.2.0] startup
[41.681s] [ext: omni.isaac.lula-1.1.0] startup
[41.694s] [ext: omni.isaac.surface_gripper-0.1.2] startup
[41.701s] [ext: omni.kit.graph.editor.core-1.3.3] startup
[41.705s] [ext: omni.isaac.motion_generation-3.6.1] startup
[41.714s] [ext: omni.isaac.manipulators-1.0.1] startup
[41.717s] [ext: omni.isaac.core_nodes-0.13.0] startup
[41.734s] [ext: omni.kit.graph.widget.variables-2.0.2] startup
[41.737s] [ext: omni.kit.graph.delegate.modern-1.6.0] startup
[41.740s] [ext: omni.isaac.franka-0.3.0] startup
[41.741s] [ext: omni.isaac.wheeled_robots-0.5.8] startup
[41.766s] [ext: omni.graph.window.core-1.23.4] startup
[41.778s] [ext: omni.graph.instancing-1.1.4] startup
[41.784s] [ext: omni.graph.tutorials-1.1.2] startup
[41.808s] [ext: omni.graph.window.action-1.3.8] startup
[41.810s] [ext: omni.isaac.isaac_sensor-1.0.2] startup
2022-12-09 08:55:09 [41,811ms] [Warning] [omni.physx.plugin] Deprecated: getSimulationEventStream is deprecated, please use getSimulationEventStreamV2
[41.825s] [ext: omni.kit.selection-0.1.0] startup
[41.827s] [ext: omni.kit.widget.layers-1.5.17] startup
[41.850s] [ext: omni.graph.bundle.action-1.0.0] startup
[41.850s] [ext: omni.kit.profiler.window-1.4.4] startup
[41.863s] [ext: omni.kit.menu.edit-1.0.6] startup
[41.866s] [ext: omni.isaac.range_sensor-0.4.3] startup
[41.897s] [ext: omni.kit.property.layer-1.1.2] startup
[41.900s] [ext: omni.replicator.isaac-1.3.2] startup
[41.922s] [ext: omni.isaac.kit-0.2.1] startup
[41.924s] [ext: omni.graph.window.generic-1.3.8] startup
[41.926s] [ext: omni.isaac.utils-0.1.11] startup
[41.931s] [ext: omni.kit.property.audio-1.0.5] startup
[41.933s] [ext: omni.kit.property.skel-1.0.1] startup
[41.936s] [ext: omni.kit.property.render-1.1.0] startup
[41.937s] [ext: omni.kit.property.camera-1.0.3] startup
[41.939s] [ext: omni.kit.property.geometry-1.2.0] startup
[41.942s] [ext: omni.kit.property.light-1.0.5] startup
[41.944s] [ext: omni.kit.property.transform-1.0.2] startup
[41.948s] [ext: omni.isaac.universal_robots-0.3.0] startup
[41.950s] [ext: omni.isaac.occupancy_map-0.2.4] startup
[41.963s] [ext: omni.kit.window.console-0.2.0] startup
[41.969s] [ext: omni.kit.window.status_bar-0.1.1] startup
[41.974s] [ext: omni.kit.property.bundle-1.2.4] startup
[41.977s] [ext: omni.kit.menu.file-1.0.8] startup
[41.978s] [ext: omni.isaac.dofbot-0.3.0] startup
[41.980s] [ext: omni.kit.streamsdk.plugins-0.0.0] startup
[41.980s] [ext: omni.kit.livestream.core-0.0.0] startup
[41.985s] [ext: omni.kit.window.title-1.1.2] startup
2022-12-09 08:55:10 [41,979ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:10 [41,979ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
2022-12-09 08:55:10 [41,980ms] [Error] [carb.windowing-glfw.plugin] GLFW initialization failed.
2022-12-09 08:55:10 [41,980ms] [Error] [carb] Failed to startup plugin carb.windowing-glfw.plugin (interfaces: [carb::windowing::IGLContext v1.0],[carb::windowing::IWindowing v1.3]) (impl: carb.windowing-glfw.plugin)
[41.989s] [ext: omni.isaac.urdf-0.4.0] startup
[42.023s] [ext: omni.kit.widget.live-0.1.0] startup
2022-12-09 08:55:10 [42,018ms] [Warning] [omni.ext-live.plugin] Unable to detect Omniverse Cache Server. File /root/.nvidia-omniverse/config/Cache/Omniverse.toml is not found. Consider installing it (version >= 280) for better IO performance.
[42.027s] [ext: omni.kit.menu.common-1.0.0] startup
[42.030s] [ext: omni.kit.livestream.native-0.1.4] startup

Active user not found. Using default user [kiosk]Streaming server started.
[42.118s] [ext: omni.isaac.sim.python-2022.1.1] startup
[42.120s] Simulation App Starting
[42.935s] app ready
[46.282s] Simulation App Startup Complete
task_name: ${task.name}
experiment: 
num_envs: 
seed: 42
torch_deterministic: False
max_iterations: 
physics_engine: physx
pipeline: gpu
sim_device: gpu
device_id: 0
rl_device: cuda:0
num_threads: 4
solver_type: 1
test: False
checkpoint: 
headless: False
wandb_activate: False
wandb_group: 
wandb_name: test
wandb_entity: wazzup
wandb_project: isaactest
task: 
    name: FrankaCabinet
    physics_engine: ${..physics_engine}
    env: 
        numEnvs: 2
        envSpacing: 3.0
        episodeLength: 500
        enableDebugVis: False
        clipObservations: 5.0
        clipActions: 1.0
        controlFrequencyInv: 2
        startPositionNoise: 0.0
        startRotationNoise: 0.0
        numProps: 4
        aggregateMode: 3
        actionScale: 7.5
        dofVelocityScale: 0.1
        distRewardScale: 2.0
        rotRewardScale: 0.5
        aroundHandleRewardScale: 10.0
        openRewardScale: 7.5
        fingerDistRewardScale: 100.0
        actionPenaltyScale: 0.01
        fingerCloseRewardScale: 10.0
    sim: 
        dt: 0.0083
        use_gpu_pipeline: True
        gravity: [0.0, 0.0, -9.81]
        add_ground_plane: True
        use_flatcache: True
        enable_scene_query_support: False
        enable_cameras: False
        default_physics_material: 
            static_friction: 1.0
            dynamic_friction: 1.0
            restitution: 0.0
        physx: 
            worker_thread_count: 4
            solver_type: 1
            use_gpu: True
            solver_position_iteration_count: 12
            solver_velocity_iteration_count: 1
            contact_offset: 0.005
            rest_offset: 0.0
            bounce_threshold_velocity: 0.2
            friction_offset_threshold: 0.04
            friction_correlation_distance: 0.025
            enable_sleeping: True
            enable_stabilization: True
            max_depenetration_velocity: 1000.0
            gpu_max_rigid_contact_count: 524288
            gpu_max_rigid_patch_count: 33554432
            gpu_found_lost_pairs_capacity: 524288
            gpu_found_lost_aggregate_pairs_capacity: 262144
            gpu_total_aggregate_pairs_capacity: 1048576
            gpu_max_soft_body_contacts: 1048576
            gpu_max_particle_contacts: 1048576
            gpu_heap_capacity: 33554432
            gpu_temp_buffer_capacity: 16777216
            gpu_max_num_partitions: 8
        franka: 
            override_usd_defaults: False
            fixed_base: False
            enable_self_collisions: False
            enable_gyroscopic_forces: True
            solver_position_iteration_count: 12
            solver_velocity_iteration_count: 1
            sleep_threshold: 0.005
            stabilization_threshold: 0.001
            density: -1
            max_depenetration_velocity: 1000.0
            contact_offset: 0.005
            rest_offset: 0.0
        cabinet: 
            override_usd_defaults: False
            fixed_base: False
            enable_self_collisions: False
            enable_gyroscopic_forces: True
            solver_position_iteration_count: 12
            solver_velocity_iteration_count: 1
            sleep_threshold: 0.0
            stabilization_threshold: 0.001
            density: -1
            max_depenetration_velocity: 1000.0
            contact_offset: 0.005
            rest_offset: 0.0
        prop: 
            override_usd_defaults: False
            fixed_base: False
            enable_self_collisions: False
            enable_gyroscopic_forces: True
            solver_position_iteration_count: 12
            solver_velocity_iteration_count: 1
            sleep_threshold: 0.005
            stabilization_threshold: 0.001
            density: 100
            max_depenetration_velocity: 1000.0
            contact_offset: 0.005
            rest_offset: 0.0
train: 
    params: 
        seed: ${...seed}
        algo: 
            name: a2c_continuous
        model: 
            name: continuous_a2c_logstd
        network: 
            name: actor_critic
            separate: False
            space: 
                continuous: 
                    mu_activation: None
                    sigma_activation: None
                    mu_init: 
                        name: default
                    sigma_init: 
                        name: const_initializer
                        val: 0
                    fixed_sigma: True
            mlp: 
                units: [256, 128, 64]
                activation: elu
                d2rl: False
                initializer: 
                    name: default
                regularizer: 
                    name: None
        load_checkpoint: False
        load_path: None
        config: 
            name: FrankaCabinet
            full_experiment_name: test
            env_name: rlgpu
            device: cuda:0
            device_name: cuda:0
            ppo: True
            mixed_precision: False
            normalize_input: True
            normalize_value: True
            num_actors: 2
            reward_shaper: 
                scale_value: 0.01
            normalize_advantage: True
            gamma: 0.99
            tau: 0.95
            learning_rate: 5e-4
            lr_schedule: adaptive
            kl_threshold: 0.008
            score_to_win: 100000000
            max_epochs: 1500
            save_best_after: 200
            save_frequency: 100
            print_stats: True
            grad_norm: 1.0
            entropy_coef: 0.0
            truncate_grads: True
            e_clip: 0.2
            horizon_length: 16
            minibatch_size: 16
            mini_epochs: 8
            critic_coef: 4
            clip_value: True
            seq_len: 4
            bounds_loss_coef: 0.0001
name: FrankaCabinet
Sim params does not have attribute:  physx
Sim params does not have attribute:  franka
Sim params does not have attribute:  cabinet
Sim params does not have attribute:  prop
Pipeline:  GPU
Pipeline Device:  cuda:0
Sim Device:  GPU
Task Device: cuda:0
RL device:  cuda:0
/isaac-sim/exts/omni.isaac.ml_archive/pip_prebundle/gym/spaces/box.py:74: UserWarning: WARN: Box bound precision lowered by casting to float32
  "Box bound precision lowered by casting to {}".format(self.dtype)
[46.493s] [ext: omni.physx.flatcache-1.4.15-5.1] startup
2022-12-09 08:55:14 [46,522ms] [Warning] [carb] Acquiring non optional plugin interface which is not listed as dependency: [omni::kit::IStageUpdate v1.0] (plugin: (null)), by client: omni.physx.flatcache.plugin. Add it to CARB_PLUGIN_IMPL_DEPS() macro of a client.
2022-12-09 08:55:14 [46,571ms] [Warning] [omni.client.plugin]  Tick: authentication: Discovery(ws://localhost:3333): Error creating Api/Connection search: Not connected
2022-12-09 08:55:15 [47,748ms] [Warning] [omni.hydra.scene_delegate.plugin] Calling getBypassRenderSkelMeshProcessing for prim /World/envs/env_0/franka/panda_rightfinger/visuals.proto_visuals_id0 that has not been populated
2022-12-09 08:55:15 [47,798ms] [Warning] [omni.physx.plugin] ParseFEMClothMaterial: UsdPrim doesn't have a PhysxSchemaPhysxDeformableSurfaceMaterialAPI

gdb output:

Thread 1 "python" received signal SIGSEGV, Segmentation fault.
0x00007fa325614917 in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/bin/libomni.physx.plugin.so
(gdb) bt
#0  0x00007fa325614917 in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/bin/libomni.physx.plugin.so
#1  0x00007fa3256e899a in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/bin/libomni.physx.plugin.so
#2  0x00007fa3256df6d8 in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/bin/libomni.physx.plugin.so
#3  0x00007fa3257f85e5 in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/bin/libomni.physx.plugin.so
#4  0x00007fa32572a9e2 in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/bin/libomni.physx.plugin.so
#5  0x00007fa325136ed6 in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/omni/physx/bindings/_physx.cpython-37m-x86_64-linux-gnu.so
#6  0x00007fa32513dde2 in ?? () from /isaac-sim/kit/extsPhysics/omni.physx-1.4.15-5.1/omni/physx/bindings/_physx.cpython-37m-x86_64-linux-gnu.so
#7  0x00000000004ba8f7 in _PyMethodDef_RawFastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:693
#8  0x00000000004ba6a6 in _PyCFunction_FastCallKeywords (func=func@entry=0x7fa3c4112fa0, args=args@entry=0x7f94c05a4a88, nargs=<optimized out>, kwnames=<optimized out>) at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:732
#9  0x00000000004c2c79 in _PyObject_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:156
#10 0x00000000004ba379 in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4619
#11 0x00000000004b6cc2 in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3093
#12 0x00000000004c4f56 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=0x7f94c05a4910) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:544
#13 function_code_fastcall (globals=<optimized out>, nargs=1, nargs@entry=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, args=<optimized out>, co=<optimized out>) at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:283
#14 _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:408
#15 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#16 0x00000000004b310b in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3110
#17 0x00000000004c4f56 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=0x7f94c059d050) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:544
#18 function_code_fastcall (globals=<optimized out>, nargs=1, nargs@entry=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, args=<optimized out>, co=<optimized out>) at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:283
#19 _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:408
#20 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#21 0x00000000004b310b in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3110
#22 0x00000000004c4f56 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=0x7f94c0fbd810) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:544
#23 function_code_fastcall (globals=<optimized out>, nargs=1, nargs@entry=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, args=<optimized out>, co=<optimized out>) at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:283
#24 _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:408
#25 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#26 0x00000000004b6cc2 in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3093
#27 0x00000000004b1411 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:547
#28 _PyEval_EvalCodeWithName () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3930
#29 0x00000000004c50ec in _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:433
#30 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#31 0x00000000004b3c41 in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3139
#32 0x00000000004b1411 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:547
#33 _PyEval_EvalCodeWithName () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3930
#34 0x00000000004c50ec in _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:433
#35 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#36 0x00000000004b310b in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3110
#37 0x00000000004b1411 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:547
#38 _PyEval_EvalCodeWithName () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3930
#39 0x00000000004c50ec in _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:433
#40 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#41 0x00000000004b6cc2 in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3093
#42 0x00000000004b1411 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:547
#43 _PyEval_EvalCodeWithName () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3930
#44 0x00000000004c50ec in _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:433
#45 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#46 0x00000000004b3c41 in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3139
#47 0x00000000004b1411 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:547
#48 _PyEval_EvalCodeWithName () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3930
#49 0x00000000004c50ec in _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:433
#50 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#51 0x00000000004b2fe1 in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3124
#52 0x00000000004c4f56 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=0x623c580) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:544
#53 function_code_fastcall (globals=<optimized out>, nargs=0, nargs@entry=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, args=<optimized out>, co=<optimized out>) at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:283
#54 _PyFunction_FastCallKeywords () at /usr/local/src/conda/python-3.7.15/Modules/opcode_targets.h:408
---Type <return> to continue, or q <return> to quit---
#55 0x00000000004ba23f in call_function () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:4616
#56 0x00000000004b2fe1 in _PyEval_EvalFrameDefault () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3124
#57 0x00000000004b1411 in PyEval_EvalFrameEx (throwflag=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, f=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:547
#58 _PyEval_EvalCodeWithName () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3930
#59 0x00000000004b1209 in PyEval_EvalCodeEx () at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:3959
#60 0x000000000054f39b in PyEval_EvalCode (co=co@entry=0x7fa8effb0ed0, globals=globals@entry=0x7fa8f0f7b370, locals=locals@entry=0x7fa8f0f7b370) at /usr/local/src/conda/python-3.7.15/Programs/gcmodule.c:524
#61 0x000000000056a263 in run_mod () at /usr/local/src/conda/python-3.7.15/Modules/Python-ast.c:1037
#62 0x0000000000570a17 in PyRun_FileExFlags () at /usr/local/src/conda/python-3.7.15/Modules/Python-ast.c:990
#63 0x000000000056ffd4 in PyRun_SimpleFileExFlags () at /usr/local/src/conda/python-3.7.15/Modules/Python-ast.c:429
#64 0x000000000056fb5b in PyRun_AnyFileExFlags () at /usr/local/src/conda/python-3.7.15/Modules/Python-ast.c:84
#65 0x0000000000544be3 in pymain_run_file (p_cf=0x7ffd63b5ade0, filename=<optimized out>, fp=0x259f7b0) at /tmp/build/80754af9/python_1669321496543/work/build-static/python.c:470
#66 pymain_run_filename (cf=0x7ffd63b5ade0, pymain=0x7ffd63b5aef0) at /tmp/build/80754af9/python_1669321496543/work/build-static/python.c:1660
#67 pymain_run_python (pymain=0x7ffd63b5aef0) at /tmp/build/80754af9/python_1669321496543/work/build-static/python.c:2970
#68 pymain_main () at /tmp/build/80754af9/python_1669321496543/work/build-static/python.c:3517
#69 0x000000000054487c in _Py_UnixMain (argc=<optimized out>, argv=<optimized out>) at /tmp/build/80754af9/python_1669321496543/work/build-static/python.c:3552
#70 0x00007fa8f0109c87 in __libc_start_main (main=0x544830 <main>, argc=5, argv=0x7ffd63b5b048, init=<optimized out>, fini=<optimized out>, rtld_fini=<optimized out>, stack_end=0x7ffd63b5b038) at ../csu/libc-start.c:310
#71 0x000000000054472e in _start () at /usr/local/src/conda/python-3.7.15/Modules/pytime.c:2515

Any help would be much appreciated.

How to get an RGB observation from a camera in the usd model???

Hi, does anyone know how to get an rgb image from the camera that has already existed in an USD model within OmniIsaacGymEnvs?

the prim_path for the camera (at a fixed location) is '/World/envs/env_0/cabinet/Camera_Prim/Camera'. I have attached a shot to my isaac sim.

I will be appreciated if anyone can help me!

watchdog: BUG: soft lockup

Hi,

The simulation and examples are working well for me, thanks a lot for that. But whenever I stop the (single-threaded) training script, I run into this warning in any other terminal on the same machine:

Message from syslogd@rlsim2 at Dec 14 09:51:09 ...
 kernel:[409187.093643] watchdog: BUG: soft lockup - CPU#54 stuck for 23s! [carb::livestrea:1460402]

and have to wait fairly long (20-30sec) for the process to shut down. Have you experienced this before or might know what is causing this?

I am running the FrankaCabinet task on an A6000 with Driver Version: 515.65.01 and a Threadripper PRO 3975WX which is far from being fully utilized. I run the scripts within the latest Isaac docker image.

Deformable physics simulation during training

Hi,

Thanks for providing this fantastic repo.

I am trying to setup a training task with deformable objects, but I found out that the deformable physics properties of the objects is not simulated during the training. But when I export the scene and simulate it in the new window, it is working. I tried to create a use.file with the deformable physics properties already set and also tried to define the deformable physics in the function set_up_scene in the task python file, but both resulted in the same issue.

Does the training environment actually support deformable physics simulation? If yes, how can I setup the correct files for that? If not, will this be considered in the future?

Thanks a lot!

PhysX error: The application needs to increase PxgDynamicsMemoryConfig::totalAggregatePairsCapacity

Hi

For the Ant environment, if the training process is not perfect (in which all ants advance and there is no overlap with other environments), several environments cross and generate the following error.

2022-06-08 12:23:10 [191,143ms] [Error] [omni.physx.plugin] PhysX error: The application needs to increase PxgDynamicsMemoryConfig::totalAggregatePairsCapacity to 4120 , otherwise, the simulation will miss interactions
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpubroadphase/src/PxgAABBManager.cpp, LINE 1064

The problem seems to be in the way of filtering collisions between clones. This error only occurs in environments where there is a possibility of intersection, but not with static environments such as Cartpole or Allegro/Shadow-Hand

By changing some training configuration (AntPPO.yaml file), it is possible to reproduce the error:

normalize_input: False
normalize_value: False
schedule_type: standard

A possible solution is to increase the gpu_total_aggregate_pairs_capacity but the most simple one is to increase the space between the environments (envSpacing)

Syntax Error

Can't solve the following syntax error. Running on Windows 11 cmd. I set PYTHON_PATH in the python.bat environment but now it's not letting me run rlgames_train.py

>>> PYTHON_PATH scripts/rlgames_train.py task=Cartpole
  File "<stdin>", line 1
    PYTHON_PATH scripts/rlgames_train.py task=Cartpole
                      ^
SyntaxError: invalid syntax

Tried running this in and out of the python environment. Any help is greatly appreciated.

[Error] [omni.hydra] UsdToMdl: Reached invalid assignment for parameter 'opacity_constant'.

When we load our own cabinet use model to train the policy, we meet some errors about omni.hydra with the following commands:

python omniisaacgymenvs/scripts/rlgames_train.py task=FrankaDrawer num_envs=1024 headless=False 

From the output we can see the training is still running, but we wonder whether these errors will influence the final results of the training, e.g., the reward or the success rates.
We really appreciate you for helping us.

The output of error is:

2023-04-23 05:25:06 [26,702ms] [Error] [omni.hydra] UsdToMdl: Reached invalid assignment for parameter 'opacity_constant'. Tried to assign a 'int'(USD) to a 'float'(MDL).

The full output is:

[2.322s] Simulation App Starting
[4.564s] app ready
[4.659s] Simulation App Startup Complete
[5.270s] [ext: omni.physx.flatcache-104.2.4-5.1] startup
Passing the following args to the base kit application:  ['task=FrankaDrawer', 'num_envs=1024', 'headless=False']
Warp 0.6.3 initialized:
   CUDA Toolkit: 11.5, Driver: 12.0
   Devices:
     "cpu"    | x86_64
     "cuda:0" | NVIDIA RTX A6000 (sm_86)
   Kernel cache: /home/rl01/.cache/warp/0.6.3
[4.614s] RTX ready
Setting seed: 42
Sim params does not have attribute:  physx
Sim params does not have attribute:  franka
Sim params does not have attribute:  cabinet
Sim params does not have attribute:  prop
Pipeline:  GPU
Pipeline Device:  cuda:0
Sim Device:  GPU
Task Device: cuda:0
RL device:  cuda:0
================= in franka drawer ===================
before set_task
in franka: tensor([0., 0., 0.]) tensor([0., 0., 0., 1.])
2023-04-23 05:25:02 [22,842ms] [Warning] [omni.hydra.scene_delegate.plugin] Calling getBypassRenderSkelMeshProcessing for prim /World/envs/env_0/franka/panda_link4/visuals.proto_visuals_id0 that has not been populated
2023-04-23 05:25:02 [23,087ms] [Warning] [carb.scenerenderer-rtx.plugin] Auto-enabling Sampled Direct Lighting because scene has 2049 lights (above 10 light threshold)
~*~*~*
~*~*~* Direct GPU Helper:
~*~*~*   2048 articulations, maxLinks=11
~*~*~*
2023-04-23 05:25:06 [26,702ms] [Error] [omni.hydra] UsdToMdl: Reached invalid assignment for parameter 'opacity_constant'. Tried to assign a 'int'(USD) to a 'float'(MDL).
2023-04-23 05:25:06 [26,704ms] [Error] [omni.hydra] UsdToMdl: Reached invalid assignment for parameter 'opacity_constant'. Tried to assign a 'int'(USD) to a 'float'(MDL).
...
2023-04-23 05:25:35 [55,957ms] [Error] [omni.hydra] UsdToMdl: Reached invalid assignment for parameter 'opacity_constant'. Tried to assign a 'int'(USD) to a 'float'(MDL).
self._drawers:<omni.isaac.core.prims.rigid_prim_view.RigidPrimView object at 0x7f28b414d810>
after set_task
Started to train
Exact experiment name requested from command line: FrankaDrawer
Box(-1.0, 1.0, (9,), float32) Box(-inf, inf, (23,), float32)
cuda:0
self.batch_size:32768, self.minibatch_size:2048
build mlp: 23
RunningMeanStd:  (1,)
RunningMeanStd:  (23,)
[2023-04-23 13:25:21] Running RL reset
fps step: 13540.9 fps step and policy inference: 13370.2  fps total: 12399.9 epoch: 1/1500
fps step: 14710.6 fps step and policy inference: 14516.9  fps total: 12807.5 epoch: 2/1500
fps step: 14519.3 fps step and policy inference: 14335.8  fps total: 13240.5 epoch: 3/1500
fps step: 13258.7 fps step and policy inference: 13102.7  fps total: 12144.6 epoch: 4/1500
fps step: 13544.6 fps step and policy inference: 13379.5  fps total: 12435.7 epoch: 5/1500
fps step: 13434.2 fps step and policy inference: 13271.2  fps total: 11878.2 epoch: 6/1500

Calculation of `rand_delta` for ShadowHand and AllegroHand

In in_hand_manipulation.py, rand_delta is calculated as follows:

rand_floats = torch_rand_float(-1.0, 1.0, (len(env_ids), self.num_hand_dofs * 2 + 5), device=self.device)
# ...
delta_max = self.hand_dof_upper_limits - self.hand_dof_default_pos
delta_min = self.hand_dof_lower_limits - self.hand_dof_default_pos
rand_delta = delta_min + (delta_max - delta_min) * rand_floats[:, 5:5+self.num_hand_dofs]
pos = self.hand_dof_default_pos + self.reset_dof_pos_noise * rand_delta

Let $L$ be delta_min and $U$ be delta_max. The line below maps rand_floats from $[-1,1]$ to $[2L-U,U]$:

rand_delta = delta_min + (delta_max - delta_min) * rand_floats[:, 5:5+self.num_hand_dofs]

However, rand_delta should be in $[L,U]$ instead. If we want to map rand_floats from $[-1,1]$ to $[L,U]$, the code above should be changed into:

rand_delta = delta_min + (delta_max - delta_min) * 0.5 * (rand_floats[:, 5:5+self.num_hand_dofs] + 1.0)

or change the assignment of rand_floats instead:

rand_floats = torch_rand_float(0.0, 1.0, (len(env_ids), self.num_hand_dofs * 2 + 5), device=self.device)

The issue also exists in the IsaacGymEnvs repo: shadow_hand.py and allegro_hand.py.

raise Exception("Failed to create simulation view backend")

PC Configuration: Ubuntu 20.04, RTX 3060, RAM 64 gb, Cuda 11.4, Nvidia driver 470.141.03.

Note: For Cartpole and Ant simulation, same command works but not for Anymal.

I was trying the demo run for anymal robot and got the following error:

rror executing job with overrides: ['task=AnymalTerrain', 'num_envs=64', 'checkpoint=omniverse://localhost/NVIDIA/Assets/Isaac/2022.1/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints/anymal_terrain.pth']
Traceback (most recent call last):
  File "scripts/rlgames_play.py", line 67, in parse_hydra_configs
    task = initialize_demo(cfg_dict, env)
  File "/home/arghya/OmniIsaacGymEnvs/omniisaacgymenvs/utils/demo_util.py", line 46, in initialize_demo
    env.set_task(task=task, sim_params=sim_config.get_physics_params(), backend="torch", init_sim=init_sim)
  File "/home/arghya/OmniIsaacGymEnvs/omniisaacgymenvs/envs/vec_env_rlgames.py", line 51, in set_task
    super().set_task(task, backend, sim_params, init_sim)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.gym/omni/isaac/gym/vec_env/vec_env_base.py", line 82, in set_task
    self._world.reset()
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.core/omni/isaac/core/world/world.py", line 281, in reset
    SimulationContext.reset(self, soft=soft)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.core/omni/isaac/core/simulation_context/simulation_context.py", line 423, in reset
    SimulationContext.initialize_physics(self)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.core/omni/isaac/core/simulation_context/simulation_context.py", line 403, in initialize_physics
    self._physics_sim_view = omni.physics.tensors.create_simulation_view(self.backend)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/kit/extsPhysics/omni.physics.tensors-1.4.15-5.1/omni/physics/tensors/impl/api.py", line 11, in create_simulation_view
    raise Exception("Failed to create simulation view backend")
Exception: Failed to create simulation view backend

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/python.sh: line 40: 19204 Segmentation fault      (core dumped) $python_exe "$@" $args
There was an error running python

Here is the full terminal output:

arghya@arghya-Pulse-GL66-12UEK:~/OmniIsaacGymEnvs/omniisaacgymenvs$ PYTHON_PATH scripts/rlgames_play.py task=AnymalTerrain num_envs=64 checkpoint=omniverse://localhost/NVIDIA/Assets/Isaac/2022.1/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints/anymal_terrain.pth 
/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/kit/python/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:251: UserWarning: In 'config': Defaults list is missing `_self_`. See https://hydra.cc/docs/upgrades/1.0_to_1.1/default_composition_order for more information
  warnings.warn(msg, UserWarning)
/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/kit/python/lib/python3.7/site-packages/hydra/_internal/defaults_list.py:412: UserWarning: In config: Invalid overriding of hydra/job_logging:
Default list overrides requires 'override' keyword.
See https://hydra.cc/docs/next/upgrades/1.0_to_1.1/defaults_list_override for more information.

  deprecation_warning(msg)
[Warning] [omni.isaac.kit.simulation_app] Modules: ['omniisaacgymenvs', 'omniisaacgymenvs.utils', 'omniisaacgymenvs.utils.hydra_cfg', 'omniisaacgymenvs.utils.hydra_cfg.hydra_utils', 'omniisaacgymenvs.utils.hydra_cfg.reformat', 'omniisaacgymenvs.utils.demo_util', 'omniisaacgymenvs.utils.config_utils', 'omniisaacgymenvs.utils.config_utils.path_utils', 'omniisaacgymenvs.envs', 'omniisaacgymenvs.envs.vec_env_rlgames', 'omni.isaac.gym', 'omni.isaac.gym.vec_env', 'omni.isaac.gym.vec_env.vec_env_base', 'omni.isaac.gym.vec_env.vec_env_mt', 'omniisaacgymenvs.scripts', 'omniisaacgymenvs.scripts.rlgames_train', 'omniisaacgymenvs.utils.rlgames', 'omniisaacgymenvs.utils.rlgames.rlgames_utils', 'omniisaacgymenvs.utils.task_util'] were loaded before SimulationApp was started and might not be loaded correctly.
[Warning] [omni.isaac.kit.simulation_app] Please check to make sure no extra omniverse or pxr modules are imported before the call to SimulationApp(...)
Starting kit application with the fillowing args:  ['/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.kit/omni/isaac/kit/simulation_app.py', '/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/apps/omni.isaac.sim.python.kit', '--/app/tokens/exe-path=/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/kit', '--/persistent/app/viewport/displayOptions=3094', '--/rtx/materialDb/syncLoads=True', '--/rtx/hydra/materialSyncLoads=True--/omni.kit.plugin/syncUsdLoads=True', '--/app/renderer/resolution/width=1280', '--/app/renderer/resolution/height=720', '--/app/window/width=1440', '--/app/window/height=900', '--/renderer/multiGpu/enabled=True', '--ext-folder', '/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts', '--ext-folder', '/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/apps', '--/physics/cudaDevice=0', '--portable']
Passing the following args to the base kit application:  ['task=AnymalTerrain', 'num_envs=64', 'checkpoint=omniverse://localhost/NVIDIA/Assets/Isaac/2022.1/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints/anymal_terrain.pth']
[Warning] [omni.kit.app.plugin] No crash reporter present, dumps uploading isn't available.
[Info] [carb] Logging to file: /home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/kit/logs/Kit/Isaac-Sim/2022.1/kit_20220918_195809.log
2022-09-19 00:58:09 [4ms] [Warning] [omni.ext.plugin] [ext: omni.drivesim.sensors.nv.lidar] Extensions config 'extension.toml' doesn't exist '/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.drivesim.sensors.nv.lidar' or '/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.drivesim.sensors.nv.lidar/config'
[0.146s] [ext: omni.stats-0.0.0] startup
[0.156s] [ext: omni.gpu_foundation-0.0.0] startup
2022-09-19 00:58:09 [155ms] [Warning] [carb] FrameworkImpl::setDefaultPlugin(client: omni.gpu_foundation_factory.plugin, desc : [carb::graphics::Graphics v2.5], plugin : carb.graphics-vulkan.plugin) failed. Plugin selection is locked, because the interface was previously acquired by: 
[0.162s] [ext: carb.windowing.plugins-1.0.0] startup
[0.168s] [ext: omni.assets.plugins-0.0.0] startup
[0.169s] [ext: omni.kit.renderer.init-0.0.0] startup
MESA-INTEL: warning: Performance support disabled, consider sysctl dev.i915.perf_stream_paranoid=0

2022-09-19 00:58:10 [1,588ms] [Warning] [gpu.foundation.plugin] Raytracing is not supported on GPU: 1. Skipping this device.

|---------------------------------------------------------------------------------------------|
| Driver Version: 470.141.3     | Graphics API: Vulkan
|=============================================================================================|
| GPU | Name                             | Active | LDA | GPU Memory | Vendor-ID | LUID       |
|     |                                  |        |     |            | Device-ID | UUID       |
|---------------------------------------------------------------------------------------------|
| 0   | NVIDIA GeForce RTX 3060 Laptop.. | Yes: 0 |     | 6144    MB | 10de      | 0          |
|     |                                  |        |     |            | 2520      | f67dad92.. |
|---------------------------------------------------------------------------------------------|
| 1   | Intel(R) Graphics (ADL GT2)      |        |     | 48011   MB | 8086      | 0          |
|     |                                  |        |     |            | 46a6      | 42f3bec9.. |
|=============================================================================================|
| OS: Linux arghya-Pulse-GL66-12UEK, Version: 5.15.0-46-generic
| XServer Vendor: The X.Org Foundation, XServer Version: 12013000 (1.20.13.0)
| Processor: 12th Gen Intel(R) Core(TM) i7-12700H | Cores: Unknown | Logical: 20
|---------------------------------------------------------------------------------------------|
| Total Memory (MB): 64014 | Free Memory: 57490
| Total Page/Swap (MB): 9536 | Free Page/Swap: 9536
|---------------------------------------------------------------------------------------------|
2022-09-19 00:58:10 [1,692ms] [Warning] [gpu.foundation.plugin] Disabling OpenGL Interop. Device 0 mismatched with OpenGL device: Mesa Intel(R) Graphics (ADL GT2).
OpenGL can only run on the main GPU. It requires the app to run with activeDevice setting set to that GPU. On laptops, you may also need to specify which GPU or driver to run on in your system (Integrated or Discrete).
2022-09-19 00:58:10 [1,693ms] [Warning] [gpu.foundation.plugin] Realm: no OpenGL interop context.
2022-09-19 00:58:11 [1,739ms] [Warning] [carb.cudainterop.plugin] On Linux only, CUDA and the display driver does not support IOMMU-enabled bare-metal PCIe peer to peer memory copy.
However, CUDA and the display driver does support IOMMU via VM pass through. As a consequence, users on Linux,
when running on a native bare metal system, should disable the IOMMU. The IOMMU should be enabled and the VFIO driver
be used as a PCIe pass through for virtual machines.
[1.754s] [ext: omni.kit.pipapi-0.0.0] startup
[1.760s] [ext: omni.kit.pip_archive-0.0.0] startup
[1.763s] [ext: omni.isaac.core_archive-1.2.0] startup
[1.775s] [ext: omni.usd.config-1.0.0] startup
[1.777s] [ext: omni.usd.libs-1.0.0] startup
[1.877s] [ext: omni.kit.pip_torch-1_11_0-0.1.3] startup
[1.897s] [ext: omni.isaac.ml_archive-1.1.0] startup
[1.898s] [ext: omni.kit.loop-isaac-0.1.0] startup
[1.899s] [ext: omni.kit.async_engine-0.0.0] startup
[1.900s] [ext: omni.appwindow-1.0.0] startup
[1.904s] [ext: omni.client-0.1.0] startup
[1.911s] [ext: omni.kit.test-0.0.0] startup
[1.911s] [ext: omni.kit.renderer.core-0.0.0] startup
[1.985s] [ext: omni.ui-2.10.3] startup
[1.999s] [ext: carb.audio-0.1.0] startup
[2.021s] [ext: omni.kit.mainwindow-0.0.0] startup
[2.022s] [ext: omni.uiaudio-1.0.0] startup
[2.024s] [ext: omni.kit.uiapp-0.0.0] startup
[2.024s] [ext: omni.usd.schema.physics-1.0.0] startup
[2.062s] [ext: omni.usd.schema.audio-0.0.0] startup
[2.068s] [ext: omni.usd.schema.semantics-0.0.0] startup
[2.076s] [ext: omni.usd.schema.omnigraph-1.0.0] startup
[2.082s] [ext: omni.usd.schema.anim-0.0.0] startup
[2.110s] [ext: omni.kit.commands-1.2.2] startup
[2.113s] [ext: omni.timeline-1.0.2] startup
[2.115s] [ext: omni.hydra.scene_delegate-0.2.0] startup
[2.119s] [ext: omni.kit.audiodeviceenum-1.0.0] startup
[2.121s] [ext: omni.usd-1.5.3] startup
[2.448s] [ext: omni.kit.asset_converter-1.2.31] startup
[2.471s] [ext: omni.usd.schema.physx-0.0.0] startup
[2.529s] [ext: omni.usd.schema.isaac-0.2.0] startup
[2.556s] [ext: omni.usd.schema.forcefield-0.0.0] startup
[2.562s] [ext: omni.kvdb-0.0.0] startup
[2.564s] [ext: omni.usdphysics-1.4.15] startup
[2.566s] [ext: omni.graph.tools-1.4.0] startup
[2.580s] [ext: omni.localcache-0.0.0] startup
[2.582s] [ext: omni.kit.stage_templates-1.1.2] startup
[2.583s] [ext: omni.convexdecomposition-1.4.15] startup
[2.585s] [ext: omni.physics.tensors-0.1.0] startup
[2.591s] [ext: omni.physx-1.4.15-5.1] startup
[4.487s] [ext: omni.graph.core-2.29.1] startup
[4.491s] [ext: omni.kit.menu.utils-1.2.11] startup
[4.499s] [ext: omni.physx.tensors-0.1.0] startup
[4.503s] [ext: omni.graph-1.23.0] startup
[4.537s] [ext: omni.kit.numpy.common-0.1.0] startup
[4.539s] [ext: omni.kit.window.script_editor-1.6.2] startup
[4.547s] [ext: omni.kit.search_core-1.0.2] startup
[4.548s] [ext: omni.isaac.dynamic_control-1.1.0] startup
[4.554s] [ext: omni.kit.renderer.capture-0.0.0] startup
[4.557s] [ext: omni.kit.widget.filebrowser-2.2.27] startup
[4.563s] [ext: omni.kit.widget.path_field-2.0.3] startup
[4.563s] [ext: omni.kit.notification_manager-1.0.5] startup
[4.565s] [ext: omni.kit.widget.versioning-1.3.8] startup
[4.566s] [ext: omni.kit.widget.browser_bar-2.0.3] startup
[4.567s] [ext: omni.kit.window.popup_dialog-2.0.8] startup
[4.568s] [ext: omni.mdl.neuraylib-0.1.0] startup
[4.571s] [ext: omni.kit.window.filepicker-2.4.30] startup
OmniAssetFileFormat
[4.610s] [ext: omni.kit.menu.create-1.0.2] startup
[4.611s] [ext: omni.mdl-0.1.0] startup
[4.626s] [ext: omni.kit.window.file_importer-1.0.4] startup
[4.627s] [ext: omni.kit.window.file_exporter-1.0.4] startup
[4.627s] [ext: omni.kit.material.library-1.3.10] startup
[4.629s] [ext: omni.kit.window.drop_support-1.0.0] startup
[4.630s] [ext: omni.kit.window.file-1.3.16] startup
[4.631s] [ext: omni.kit.context_menu-1.3.9] startup
[4.634s] [ext: omni.kit.window.property-1.6.3] startup
[4.635s] [ext: omni.kit.window.content_browser-2.4.28] startup
[4.642s] [ext: omni.kit.widget.stage-2.6.15] startup
[4.646s] [ext: omni.isaac.version-1.0.0] startup
[4.646s] [ext: omni.kit.property.usd-3.14.9] startup
[4.668s] [ext: omni.kit.viewport.legacy_gizmos-1.0.0] startup
[4.671s] [ext: omni.hydra.rtx-0.1.0] startup
[4.678s] [ext: omni.renderer-rtx-0.0.0] startup
[4.678s] [ext: omni.hydra.engine.stats-1.0.0] startup
[4.683s] [ext: omni.debugdraw-0.1.0] startup
[4.687s] [ext: omni.kit.widget.settings-1.0.0] startup
[4.688s] [ext: omni.kit.window.viewport-0.0.0] startup
[7.739s] [ext: omni.kit.widget.prompt-1.0.1] startup
[7.739s] [ext: omni.kit.widget.graph-1.4.3] startup
[7.791s] [ext: omni.kit.window.preferences-1.2.1] startup
[7.814s] [ext: omni.ui_query-1.1.1] startup
[7.816s] [ext: omni.graph.ui-1.6.1] startup
[7.831s] [ext: omni.kit.ui_test-1.2.2] startup
[7.833s] [ext: omni.graph.action-1.18.0] startup
[7.841s] [ext: omni.kit.widget.searchfield-1.0.6] startup
[7.842s] [ext: omni.kit.usd_undo-0.1.0] startup
[7.843s] [ext: omni.graph.scriptnode-0.5.0] startup
[7.844s] [ext: omni.physx.commands-1.4.15-5.1] startup
[7.848s] [ext: omni.graph.nodes-1.26.0] startup
[7.857s] [ext: omni.command.usd-1.0.1] startup
[7.859s] [ext: omni.kit.window.extensions-1.1.0] startup
[7.863s] [ext: omni.syntheticdata-0.2.1] startup
[7.886s] [ext: omni.kit.primitive.mesh-1.0.0] startup
[7.888s] [ext: omni.warp-0.2.2] startup
Warp initialized:
   Version: 0.2.2
   CUDA device: NVIDIA GeForce RTX 3060 Laptop GPU
   Kernel cache: /home/arghya/.cache/warp/0.2.2
[7.974s] [ext: omni.isaac.ui-0.2.1] startup
[7.976s] [ext: omni.replicator.core-1.4.3] startup
[8.093s] [ext: omni.isaac.core-1.24.3] startup
[8.148s] [ext: omni.physx.ui-1.4.15-5.1] startup
[8.194s] [ext: omni.kit.property.material-1.8.5] startup
[8.195s] [ext: omni.kit.window.toolbar-1.2.4] startup
[8.199s] [ext: omni.isaac.core_nodes-0.13.0] startup
[8.206s] [ext: omni.physx.demos-1.4.15-5.1] startup
[8.208s] [ext: omni.kit.property.physx-0.1.0] startup
2022-09-19 00:58:17 [8,240ms] [Warning] [omni.physx.plugin] Deprecated: getSimulationEventStream is deprecated, please use getSimulationEventStreamV2
[8.247s] [ext: omni.physx.tests-1.4.15-5.1] startup
[8.452s] [ext: omni.isaac.wheeled_robots-0.5.8] startup
[8.462s] [ext: omni.kit.menu.common-1.0.0] startup
[8.464s] [ext: omni.physx.vehicle-1.4.15-5.1] startup
[8.475s] [ext: omni.physx.cct-1.4.15-5.1] startup
[8.527s] [ext: omni.physx.camera-1.4.15-5.1] startup
[8.533s] [ext: omni.kit.widget.stage_icons-1.0.2] startup
[8.534s] [ext: omni.ui.scene-1.5.0] startup
[8.539s] [ext: omni.physx.bundle-1.4.15-5.1] startup
[8.539s] [ext: omni.kit.window.stage-2.3.7] startup
[8.541s] [ext: omni.replicator.composer-1.1.3] startup
[8.544s] [ext: omni.isaac.lula-1.1.0] startup
[8.552s] [ext: omni.rtx.window.settings-0.6.1] startup
[8.557s] [ext: omni.isaac.surface_gripper-0.1.2] startup
[8.560s] [ext: omni.isaac.motion_planning-0.2.0] startup
[8.566s] [ext: omni.rtx.settings.core-0.5.5] startup
[8.569s] [ext: omni.isaac.manipulators-1.0.1] startup
[8.570s] [ext: omni.isaac.motion_generation-3.6.1] startup
[8.574s] [ext: omni.kit.widget.zoombar-1.0.3] startup
[8.574s] [ext: omni.kit.graph.delegate.default-1.0.15] startup
[8.576s] [ext: omni.isaac.franka-0.3.0] startup
[8.576s] [ext: omni.kit.browser.core-2.0.12] startup
[8.580s] [ext: omni.kit.graph.editor.core-1.3.3] startup
[8.582s] [ext: omni.kit.graph.usd.commands-1.1.0] startup
[8.582s] [ext: omni.kit.browser.folder.core-1.1.13] startup
[8.584s] [ext: omni.kit.graph.widget.variables-2.0.2] startup
[8.585s] [ext: omni.kit.graph.delegate.modern-1.6.0] startup
[8.586s] [ext: omni.kit.selection-0.1.0] startup
[8.587s] [ext: omni.isaac.debug_draw-0.1.2] startup
[8.591s] [ext: omni.graph.window.core-1.23.4] startup
[8.595s] [ext: omni.graph.instancing-1.1.4] startup
[8.600s] [ext: omni.kit.menu.edit-1.0.6] startup
[8.601s] [ext: omni.isaac.isaac_sensor-1.0.2] startup
2022-09-19 00:58:17 [8,601ms] [Warning] [omni.physx.plugin] Deprecated: getSimulationEventStream is deprecated, please use getSimulationEventStreamV2
[8.609s] [ext: omni.graph.tutorials-1.1.2] startup
[8.617s] [ext: omni.graph.window.action-1.3.8] startup
[8.619s] [ext: omni.kit.widget.live-0.1.0] startup
[8.622s] [ext: omni.kit.widget.layers-1.5.17] startup
[8.630s] [ext: omni.graph.bundle.action-1.0.0] startup
[8.630s] [ext: omni.isaac.range_sensor-0.4.3] startup
[8.647s] [ext: omni.kit.property.layer-1.1.2] startup
[8.649s] [ext: omni.replicator.isaac-1.3.2] startup
[8.657s] [ext: omni.isaac.kit-0.2.1] startup
[8.658s] [ext: omni.graph.window.generic-1.3.8] startup
[8.658s] [ext: omni.isaac.utils-0.1.11] startup
[8.661s] [ext: omni.kit.property.audio-1.0.5] startup
[8.662s] [ext: omni.kit.property.skel-1.0.1] startup
[8.663s] [ext: omni.kit.property.render-1.1.0] startup
[8.664s] [ext: omni.kit.property.camera-1.0.3] startup
[8.664s] [ext: omni.kit.property.geometry-1.2.0] startup
[8.666s] [ext: omni.kit.property.light-1.0.5] startup
[8.666s] [ext: omni.kit.property.transform-1.0.2] startup
[8.668s] [ext: omni.isaac.universal_robots-0.3.0] startup
[8.669s] [ext: omni.isaac.occupancy_map-0.2.4] startup
[8.679s] [ext: omni.kit.window.console-0.2.0] startup
[8.683s] [ext: omni.kit.window.status_bar-0.1.1] startup
[8.686s] [ext: omni.kit.property.bundle-1.2.4] startup
[8.687s] [ext: omni.kit.menu.file-1.0.8] startup
[8.688s] [ext: omni.kit.manipulator.viewport-1.0.6] startup
[8.689s] [ext: omni.isaac.urdf-0.4.0] startup
[8.708s] [ext: omni.isaac.dofbot-0.3.0] startup
[8.708s] [ext: omni.kit.window.title-1.1.2] startup
[8.709s] [ext: omni.kit.profiler.window-1.4.4] startup
[8.713s] [ext: omni.graph.visualization.nodes-1.1.1] startup
[8.716s] [ext: omni.isaac.synthetic_utils-0.3.5] startup
[8.717s] [ext: semantics.schema.editor-0.2.3] startup
[8.719s] [ext: omni.isaac.sim.python-2022.1.1] startup
[8.720s] Simulation App Starting
[9.362s] app ready
[11.049s] Simulation App Startup Complete
task: 
    name: AnymalTerrain
    physics_engine: physx
    env: 
        numEnvs: 64
        numObservations: 188
        numActions: 12
        envSpacing: 3.0
        terrain: 
            staticFriction: 1.0
            dynamicFriction: 1.0
            restitution: 0.0
            curriculum: True
            maxInitMapLevel: 0
            mapLength: 8.0
            mapWidth: 8.0
            numLevels: 10
            numTerrains: 20
            terrainProportions: [0.1, 0.1, 0.35, 0.25, 0.2]
            slopeTreshold: 0.5
        baseInitState: 
            pos: [0.0, 0.0, 0.62]
            rot: [1.0, 0.0, 0.0, 0.0]
            vLinear: [0.0, 0.0, 0.0]
            vAngular: [0.0, 0.0, 0.0]
        randomCommandVelocityRanges: 
            linear_x: [-1.0, 1.0]
            linear_y: [-1.0, 1.0]
            yaw: [-3.14, 3.14]
        control: 
            stiffness: 80.0
            damping: 2.0
            actionScale: 0.5
            decimation: 4
        defaultJointAngles: 
            LF_HAA: 0.03
            LH_HAA: 0.03
            RF_HAA: -0.03
            RH_HAA: -0.03
            LF_HFE: 0.4
            LH_HFE: -0.4
            RF_HFE: 0.4
            RH_HFE: -0.4
            LF_KFE: -0.8
            LH_KFE: 0.8
            RF_KFE: -0.8
            RH_KFE: 0.8
        learn: 
            terminalReward: 0.0
            linearVelocityXYRewardScale: 1.0
            linearVelocityZRewardScale: -4.0
            angularVelocityXYRewardScale: -0.05
            angularVelocityZRewardScale: 0.5
            orientationRewardScale: -0.0
            torqueRewardScale: -2e-05
            jointAccRewardScale: -0.0005
            baseHeightRewardScale: -0.0
            actionRateRewardScale: -0.01
            fallenOverRewardScale: -1.0
            hipRewardScale: -0.0
            linearVelocityScale: 2.0
            angularVelocityScale: 0.25
            dofPositionScale: 1.0
            dofVelocityScale: 0.05
            heightMeasurementScale: 5.0
            addNoise: True
            noiseLevel: 1.0
            dofPositionNoise: 0.01
            dofVelocityNoise: 1.5
            linearVelocityNoise: 0.1
            angularVelocityNoise: 0.2
            gravityNoise: 0.05
            heightMeasurementNoise: 0.06
            pushInterval_s: 15
            episodeLength_s: 20
    sim: 
        dt: 0.005
        up_axis: z
        use_gpu_pipeline: True
        gravity: [0.0, 0.0, -9.81]
        add_ground_plane: False
        use_flatcache: True
        enable_scene_query_support: False
        enable_cameras: False
        default_physics_material: 
            static_friction: 1.0
            dynamic_friction: 1.0
            restitution: 0.0
        physx: 
            worker_thread_count: 4
            solver_type: 1
            use_gpu: True
            solver_position_iteration_count: 4
            solver_velocity_iteration_count: 0
            contact_offset: 0.02
            rest_offset: 0.0
            bounce_threshold_velocity: 0.2
            friction_offset_threshold: 0.04
            friction_correlation_distance: 0.025
            enable_sleeping: True
            enable_stabilization: True
            max_depenetration_velocity: 100.0
            gpu_max_rigid_contact_count: 524288
            gpu_max_rigid_patch_count: 163840
            gpu_found_lost_pairs_capacity: 4194304
            gpu_found_lost_aggregate_pairs_capacity: 33554432
            gpu_total_aggregate_pairs_capacity: 4194304
            gpu_max_soft_body_contacts: 1048576
            gpu_max_particle_contacts: 1048576
            gpu_heap_capacity: 134217728
            gpu_temp_buffer_capacity: 33554432
            gpu_max_num_partitions: 8
        anymal: 
            override_usd_defaults: False
            fixed_base: False
            enable_self_collisions: True
            enable_gyroscopic_forces: False
            solver_position_iteration_count: 4
            solver_velocity_iteration_count: 0
            sleep_threshold: 0.005
            stabilization_threshold: 0.001
            density: -1
            max_depenetration_velocity: 100.0
train: 
    params: 
        seed: 42
        algo: 
            name: a2c_continuous
        model: 
            name: continuous_a2c_logstd
        network: 
            name: actor_critic
            separate: True
            space: 
                continuous: 
                    mu_activation: None
                    sigma_activation: None
                    mu_init: 
                        name: default
                    sigma_init: 
                        name: const_initializer
                        val: 0.0
                    fixed_sigma: True
            mlp: 
                units: [512, 256, 128]
                activation: elu
                d2rl: False
                initializer: 
                    name: default
                regularizer: 
                    name: None
        load_checkpoint: True
        load_path: /home/arghya/OmniIsaacGymEnvs/omniisaacgymenvs/checkpoints/anymal_terrain.pth
        config: 
            name: AnymalTerrain
            full_experiment_name: AnymalTerrain
            device: cuda:0
            device_name: cuda:0
            env_name: rlgpu
            ppo: True
            mixed_precision: False
            normalize_input: True
            normalize_value: True
            normalize_advantage: True
            value_bootstrap: True
            clip_actions: False
            num_actors: 64
            reward_shaper: 
                scale_value: 1.0
            gamma: 0.99
            tau: 0.95
            e_clip: 0.2
            entropy_coef: 0.001
            learning_rate: 0.0003
            lr_schedule: adaptive
            kl_threshold: 0.008
            truncate_grads: True
            grad_norm: 1.0
            horizon_length: 48
            minibatch_size: 16384
            mini_epochs: 5
            critic_coef: 2
            clip_value: True
            seq_len: 4
            bounds_loss_coef: 0.0
            max_epochs: 2000
            save_best_after: 100
            score_to_win: 20000
            save_frequency: 50
            print_stats: True
task_name: AnymalTerrain
experiment: 
num_envs: 64
seed: 42
torch_deterministic: False
max_iterations: 
physics_engine: physx
pipeline: gpu
sim_device: gpu
device_id: 0
rl_device: cuda:0
num_threads: 4
solver_type: 1
test: False
checkpoint: /home/arghya/OmniIsaacGymEnvs/omniisaacgymenvs/checkpoints/anymal_terrain.pth
headless: False
wandb_activate: False
wandb_group: 
wandb_name: AnymalTerrain
wandb_entity: 
wandb_project: omniisaacgymenvs
Sim params does not have attribute:  up_axis
Sim params does not have attribute:  physx
Sim params does not have attribute:  anymal
Pipeline:  GPU
Pipeline Device:  cuda:0
Sim Device:  GPU
Task Device: cuda:0
RL device:  cuda:0
/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.ml_archive/pip_prebundle/gym/spaces/box.py:74: UserWarning: WARN: Box bound precision lowered by casting to float32
  "Box bound precision lowered by casting to {}".format(self.dtype)
/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.kit.pip_torch-1_11_0-0.1.3+103.1.lx64.cp37/torch-1-11-0/torch/functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at  ../aten/src/ATen/native/TensorShape.cpp:2228.)
  return _VF.meshgrid(tensors, **kwargs)  # type: ignore[attr-defined]
[11.307s] [ext: omni.physx.flatcache-1.4.15-5.1] startup
2022-09-19 00:58:20 [11,314ms] [Warning] [carb] Acquiring non optional plugin interface which is not listed as dependency: [omni::kit::IStageUpdate v1.0] (plugin: (null)), by client: omni.physx.flatcache.plugin. Add it to CARB_PLUGIN_IMPL_DEPS() macro of a client.
2022-09-19 00:58:21 [12,357ms] [Warning] [omni.hydra.scene_delegate.plugin] Calling getBypassRenderSkelMeshProcessing for prim /World/envs/env_0/anymal/base/visuals.proto_mesh_20_id20 that has not been populated
2022-09-19 00:58:22 [12,855ms] [Warning] [omni.physx.plugin] ParseFEMClothMaterial: UsdPrim doesn't have a PhysxSchemaPhysxDeformableSurfaceMaterialAPI

PxgCudaDeviceMemoryAllocator fail to allocate memory 134217728 bytes!! Result = 2
2022-09-19 00:58:31 [21,847ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams cuEventRecord failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 54
2022-09-19 00:58:31 [21,847ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,847ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams cuStreamWaitEvent failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 60
2022-09-19 00:58:31 [21,847ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,847ms] [Error] [omni.physx.plugin] PhysX error: memcpy failed fail!
  700, FILE /buildAgent/work/99bede84aa0a52c2/source/gpunarrowphase/src/PxgNarrowphaseCore.cpp, LINE 2408
2022-09-19 00:58:31 [21,847ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams cuEventRecord failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 54
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams cuStreamWaitEvent failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 60
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams cuEventRecord failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 54
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams cuStreamWaitEvent failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 60
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 80
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 80
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveContactParallel fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1492
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveContactParallel fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1529
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU compute solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1697
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveSelfConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 859
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU artiSolveInternalTendonConstraints1T fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 879
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveInternalConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 907
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU propagate solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1784
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU compute solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1697
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveSelfConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 859
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU artiSolveInternalTendonConstraints1T fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 879
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveInternalConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 907
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU propagate solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1784
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU compute solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1697
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveSelfConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 859
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU artiSolveInternalTendonConstraints1T fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 879
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveInternalConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 907
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU propagate solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1784
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU compute solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1697
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveSelfConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 859
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU artiSolveInternalTendonConstraints1T fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 879
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU solveInternalConstraints fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpuarticulation/src/PxgArticulationCore.cpp, LINE 907
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: GPU propagate solver bodies average velocities fail to launch kernel!!
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpusolver/src/PxgTGSCudaSolverCore.cpp, LINE 1784
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 80
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,848ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 80
2022-09-19 00:58:31 [21,849ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [21,849ms] [Error] [omni.physx.plugin] PhysX error: SynchronizeStreams failed
, FILE /buildAgent/work/99bede84aa0a52c2/source/gpucommon/include/PxgCudaUtils.h, LINE 80
2022-09-19 00:58:31 [21,849ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.plugin] PhysX error: PhysX Internal CUDA error. Simulation can not continue!, FILE /buildAgent/work/99bede84aa0a52c2/source/physx/src/NpScene.cpp, LINE 3509
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.plugin] Cuda context manager error, simulation will be stopped and new cuda context manager will be created.
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 146
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA context validation failed
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 193
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] Failed to initialize GPU simulation data
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 457
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 458
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 459
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 460
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 461
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 462
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 463
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 464
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 465
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 466
2022-09-19 00:58:31 [22,049ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 467
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 468
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 469
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 471
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 472
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 474
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 475
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 476
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 478
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 479
2022-09-19 00:58:31 [22,050ms] [Error] [omni.physx.tensors.plugin] CUDA error: an illegal memory access was encountered: ../../../source/extensions/omni.physx.tensors/plugins/gpu/GpuSimulationData.cpp: 480
Error executing job with overrides: ['task=AnymalTerrain', 'num_envs=64', 'checkpoint=omniverse://localhost/NVIDIA/Assets/Isaac/2022.1/Isaac/Samples/OmniIsaacGymEnvs/Checkpoints/anymal_terrain.pth']
Traceback (most recent call last):
  File "scripts/rlgames_play.py", line 67, in parse_hydra_configs
    task = initialize_demo(cfg_dict, env)
  File "/home/arghya/OmniIsaacGymEnvs/omniisaacgymenvs/utils/demo_util.py", line 46, in initialize_demo
    env.set_task(task=task, sim_params=sim_config.get_physics_params(), backend="torch", init_sim=init_sim)
  File "/home/arghya/OmniIsaacGymEnvs/omniisaacgymenvs/envs/vec_env_rlgames.py", line 51, in set_task
    super().set_task(task, backend, sim_params, init_sim)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.gym/omni/isaac/gym/vec_env/vec_env_base.py", line 82, in set_task
    self._world.reset()
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.core/omni/isaac/core/world/world.py", line 281, in reset
    SimulationContext.reset(self, soft=soft)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.core/omni/isaac/core/simulation_context/simulation_context.py", line 423, in reset
    SimulationContext.initialize_physics(self)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/exts/omni.isaac.core/omni/isaac/core/simulation_context/simulation_context.py", line 403, in initialize_physics
    self._physics_sim_view = omni.physics.tensors.create_simulation_view(self.backend)
  File "/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/kit/extsPhysics/omni.physics.tensors-1.4.15-5.1/omni/physics/tensors/impl/api.py", line 11, in create_simulation_view
    raise Exception("Failed to create simulation view backend")
Exception: Failed to create simulation view backend

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
/home/arghya/.local/share/ov/pkg/isaac_sim-2022.1.1/python.sh: line 40: 19204 Segmentation fault      (core dumped) $python_exe "$@" $args
There was an error running python

Ane help is appreciated. TIA.

Different behaviour from isaac sim

Hello,
I have a pick and place application, where a robot is grasping a cloth created using ClothPrim.
I started from isaac sim and i was able to grasp the cloth, and when i integrated everything inside this repository i have the following behavior.

The cloth parameters are exactly the same and i use the same robot model as well. So i don't know exactly what to do...

Screencast.from.04-21-2023.12.02.09.PM.webm

domain randomization seems CANNOT change rigid prims' material properties

Hi,

I have tried the provided domain randomization code for the shadow hand task. I want to apply noises to object_view's material properties, especially restitution, on reset. I randomize the restitution of objects in two ranges, [0.0, 0.1] or [0.9, 0.99], and launched two runs. However, the behavior of objects dropping to the ground seems identical.

More specifically, I have made the following modifications to the released code:

• In the domain_randomization/randomization_params/rigid_prim_views/object_view/material_properties/on_reset of the ShadowHand task config:
  (1) set the operation to direct;
  (2) set the distribution parameters to [[0.7, 1, 0.0], [1.3, 1, 0.1]] or [[0.7, 1, 0.9], [1.3, 1, 0.99]].
• In the task/shared/in_hand_manipulation.py, I register a GeometryPrimView for the ground plane, and apply a physics material with restitution 0.5 to it, during set_up_scene.

I launched two runs with the two different distribution parameters (one with restitution in [0.0, 0.1], and the other with restitution in [0.9, 0.99]). I printed object_view._physics_view.get_material_properties(), which is indeed some random value in [0.0, 0.1] or [0.9, 0.99], but the behavior of the simulation environment seems identical for these two different settings of object restitutions. When objects drop to the ground from the hand, none of these two restitution ranges makes the object bounce more than the other.

Why are there identical behaviors under different restitution ranges? I have attached the modified code below. Is there something wrong with it?

omniisaacgymenvs_restitution_dr.zip

6x slower compared to IsaacGymEnvs

Why is OmniIsaacGymEnvs (Omniverse Isaac Sim) version about 6x slower, compared with IsaacGymEnvs (IsaacGym Preview) version, with same hardware?

Tested on Humanoid environment, headless, with 8192 environments (RTX 3080 Laptop)

IsaacGymEnvs TPS: ~260k
OmniIsaacGymEnvs TPS: ~40k

It is also observed that OmniIsaacGymEnvs have only 5% GPU utilization, while IsaacGymEnvs reaches 70% GPU utilization

Test code for OmniIsaacGymEnvs

PYTHON_PATH scripts/random_policy.py task=Humanoid headless=True task.env.numEnvs=8192

(random_policy.py modified, added a TPS counter like below)

Test code for IsaacGymEnvs

import isaacgym
import isaacgymenvs
import torch

import time


device = "cuda:0"
env_num = 8192
total_steps = int(1e6)

envs = isaacgymenvs.make(
    seed=0, 
    task="Humanoid", 
    num_envs=env_num, 
    sim_device=device,
    rl_device=device,
    headless=True
)
print("Observation space is", envs.observation_space)
print("Action space is", envs.action_space)


last_time = time.time()
last_steps = 0

obs = envs.reset()
for _ in range(total_steps):
    obs, reward, done, info = envs.step(torch.rand((env_num,)+envs.action_space.shape, device=device))

    last_steps += 1
    if last_steps % 100 == 0:
        cur_time = time.time()
        tps = last_steps * env_num / (cur_time - last_time)
        
        last_time = cur_time
        last_steps = 0

        print ("TPS {:.1f}".format(tps))

error when running rlgames_train

Hi all,

Thanks for your awesome contribution.

When I'm trying to play the demo, it occurs the problem for both rlgames_train and rlgames_train_mt as:
$ PYTHON_PATH scripts/rlgames_train_mt.py task=Ant headless=True
/home/USER/.local/share/ov/pkg/isaac_sim-2022.1.0/python.sh: line 46: 60079 Segmentation fault (core dumped) $python_exe "$@" $args There was an error running python

I tried to build conda environment as well, but the same Segmentation fault error comes.

btw, random_policy works well on my side.
PYTHON_PATH -m pip install -e . is done as well.

Desktop: Ubuntu 18.04 + RTX 3080 GPU

Appreciate it if you can give some hints.
Fan

Video recording during train

Does this platform support video recording during training? This function is supported in IsaacGymEnv. Thanks!

ModuleNotFoundError: No module named 'omni.isaac.franka'

This error occurs when I tried to run the FrankaCabinet taks with headless mode. I didn't change the script under tasks/franka_cabinet.py. The command is as follows:

/home/midea/.local/share/ov/pkg/isaac_sim-*/python.sh omniisaacgymenvs/scripts/rlgames_train.py task=FrankaCabinet headless=True

Then I got:

[1.760s] [ext: omni.kit.window.title-1.1.2] startup
[1.761s] [ext: omni.isaac.gym-0.3.3] startup
[1.761s] [ext: omni.isaac.sim.python.gym.headless-2022.2.1] startup
[1.762s] Simulation App Starting
[2.268s] app ready
2023-04-19 12:38:12 [2,265ms] [Warning] [omni.client.plugin]  HTTP Client: provider_http: CC-493: Request through cache failed. Retrying without cache for http://omniverse-content-production.s3-us-west-2.amazonaws.com/.cloudfront.toml
2023-04-19 12:38:12 [2,265ms] [Warning] [omni.client.plugin]  HTTP Client: omniclient: CC-873: Bypassing cache until the application is restarted
[2.284s] Simulation App Startup Complete
Setting seed: 42
Sim params does not have attribute:  physx
Sim params does not have attribute:  franka
Sim params does not have attribute:  cabinet
Sim params does not have attribute:  prop
Pipeline:  GPU
Pipeline Device:  cuda:0
Sim Device:  GPU
[2.309s] [ext: omni.inspect-1.0.1] startup
[2.316s] [ext: omni.kit.clipboard-1.0.0] startup
[2.320s] [ext: omni.kit.menu.create-1.0.8] startup
[2.327s] [ext: omni.volume-0.1.0] startup
[2.330s] [ext: omni.kit.context_menu-1.5.12] startup
[2.333s] [ext: omni.activity.core-1.0.1] startup
[2.335s] [ext: omni.hydra.rtx-0.1.0] startup
[2.342s] [ext: omni.debugdraw-0.1.1] startup
[2.348s] [ext: omni.kit.widget.stage-2.7.24] startup
[2.352s] [ext: omni.kit.window.property-1.8.2] startup
[2.354s] [ext: omni.kit.viewport.utility-1.0.14] startup
[2.355s] [ext: omni.kit.property.usd-3.18.17] startup
[2.362s] [ext: omni.kit.widget.text_editor-1.0.2] startup
[2.364s] [ext: omni.kit.widget.settings-1.0.1] startup
[2.365s] [ext: omni.kit.widget.graph-1.5.6-104_2] startup
[2.377s] [ext: omni.ui.scene-1.5.18] startup
[2.382s] [ext: omni.kit.window.preferences-1.3.8] startup
[2.480s] [ext: omni.kit.window.extensions-1.1.1] startup
[2.484s] [ext: omni.kit.widget.prompt-1.0.5] startup
[2.485s] [ext: omni.graph.ui-1.24.2] startup
[2.519s] [ext: omni.graph.scriptnode-0.10.0] startup
[2.521s] [ext: omni.graph.action-1.31.1] startup
[2.530s] [ext: omni.graph.bundle.action-1.3.0] startup
[2.530s] [ext: omni.syntheticdata-0.2.4] startup
2023-04-19 12:38:12 [2,527ms] [Warning] [omni.syntheticdata.scripts.extension] SyntheticData extension needs at least a stageFrameHistoryCount of 3
[2.546s] [ext: omni.command.usd-1.0.2] startup
[2.548s] [ext: omni.replicator.core-1.7.8] startup
2023-04-19 12:38:12 [2,553ms] [Warning] [omni.replicator.core.scripts.annotators] Annotator PostProcessDispatch is already registered, overwriting annotator template
[2.616s] [ext: omni.replicator.isaac-1.7.4] startup
Error executing job with overrides: ['task=FrankaCabinet', 'num_envs=1', 'headless=True']
2023-04-19 12:38:12 [2,822ms] [Warning] [omni.client.plugin]  HTTP Client: provider_http: CC-493: Request through cache failed. Retrying without cache for http://omniverse-content-production.s3-us-west-2.amazonaws.com/
An error occurred during Hydra's exception formatting:
AssertionError()
Traceback (most recent call last):
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py", line 252, in run_and_report
    assert mdl is not None
AssertionError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "omniisaacgymenvs/scripts/rlgames_train.py", line 143, in <module>
    parse_hydra_configs()
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/main.py", line 52, in decorated_main
    config_name=config_name,
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py", line 378, in _run_hydra
    lambda: hydra.run(
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py", line 294, in run_and_report
    raise ex
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py", line 211, in run_and_report
    return func()
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/utils.py", line 381, in <lambda>
    overrides=args.overrides,
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/_internal/hydra.py", line 111, in run
    _ = ret.return_value
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/core/utils.py", line 233, in return_value
    raise self._return_value
  File "/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/kit/python/lib/python3.7/site-packages/hydra/core/utils.py", line 160, in run_job
    ret.return_value = task_function(task_cfg)
  File "omniisaacgymenvs/scripts/rlgames_train.py", line 115, in parse_hydra_configs
    task = initialize_task(cfg_dict, env)
  File "/home/midea/wk/isaac_proj/OmniIsaacGymEnvs/omniisaacgymenvs/utils/task_util.py", line 41, in initialize_task
    from omniisaacgymenvs.tasks.franka_drawer import FrankaDrawerTask
  File "/home/midea/wk/isaac_proj/OmniIsaacGymEnvs/omniisaacgymenvs/tasks/franka_drawer.py", line 33, in <module>
    from omni.isaac.franka import Franka
ModuleNotFoundError: No module named 'omni.isaac.franka'
2023-04-19 12:38:13 [3,097ms] [Warning] [carb.audio.context] 1 contexts were leaked
/home/midea/.local/share/ov/pkg/isaac_sim-2022.2.1/python.sh: line 41: 10311 Segmentation fault      (core dumped) $python_exe "$@" $args
There was an error running python

But this error gone when I set headless=False. How can I use headless mode with the task FrankaCabinet?

Domain Randomization

Hi,

The standalone Isaac Gym provides a convenient domain randomization method by adding some parameters in config files. But for Omniverse Gym, I currently cannot find a method of doing so. Is there any workaround to achieve randomizations like adding noises to observations/states, and randomizing the friction/mass of an object?

Thanks!

How to generate videos of training ?

Hello, I've been looking around the doc and examples for a while but I can't find an example. Is there a way to generate videos, either during training at some points, or at the end of training ?

I've seen that contrary to the original IsaacGym repo, the render "rgb" function has not been implemented here, but on the other hand there's a very recent commit called "d62b283 support headless render" which would be exactly what I want.

Even just outputing a single image would be fine, I could do the script later to make a video.

Thanks.

Issue when running hand manipulation tasks

Hi,

Thanks for your amazing work!

I'm new to Isaac Sim, and I'm trying to train/test the ShadowHand and AllegroHand tasks on my Windows machine with the default parameters using the example commands:

PYTHON_PATH scripts/rlgames_train.py task=ShadowHand and PYTHON_PATH scripts/random_policy.py task=ShadowHand

However, I always got the following warnings and nothing shows on Isaac Sim UI:

...
Pipeline: GPU
Pipeline Device: cuda:0
Sim Device: GPU
Obs type: full
Task Device: cuda:0
RL device: cuda:0
C:\Research\Omniverse\ov\pkg\isaac_sim-2022.1.1/exts/omni.isaac.ml_archive/pip_prebundle\gym\spaces\box.py:74: UserWarning: WARN: Box bound precision lowered by casting to float32
"Box bound precision lowered by casting to {}".format(self.dtype)
[15.269s] [ext: omni.physx.flatcache-1.4.15-5.1] startup
2022-10-13 01:01:26 [15,426ms] [Warning] [carb] Acquiring non optional plugin interface which is not listed as dependency: [omni::kit::IStageUpdate v1.0] (plugin: (null)), by client: omni.physx.flatcache.plugin. Add it to CARB_PLUGIN_IMPL_DEPS() macro of a client.
2022-10-13 01:01:29 [18,563ms] [Warning] [omni.client.plugin] Tick: authentication: Discovery(ws://localhost:3333): Auth/Credentials service not found!
2022-10-13 01:01:41 [30,303ms] [Warning] [omni.hydra.scene_delegate.plugin] Calling getBypassRenderSkelMeshProcessing for prim /World/envs/env_0/shadow_hand/robot0_mfproximal/visuals.proto_robot0_V_mfproximal_id0 that has not been populated
2022-10-13 01:01:42 [31,566ms] [Warning] [omni.physx.plugin] ParseFEMClothMaterial: UsdPrim doesn't have a PhysxSchemaPhysxDeformableSurfaceMaterialAPI

2022-10-13 01:01:44 [32,999ms] [Warning] [omni.usd] Coding Error: in _ValidateEditPrim at line 1279 of C:\b\w\ca6c508eae419cf8\USD\pxr\usd\usd\stage.cpp -- Cannot create prim spec at path </__Master_4/robot0_C_forearm>; authoring to an instancing master is not allowed.

2022-10-13 01:01:44 [33,001ms] [Warning] [omni.usd] Warning: in AddAppliedSchema at line 300 of C:\b\w\ca6c508eae419cf8\USD\pxr\usd\usd\prim.cpp -- Unable to create primSpec at path </__Master_4/robot0_C_forearm> in edit target 'anon:00000189A4CFC0E0:World0.usd'. Failed to add applied API schema.

2022-10-13 01:01:44 [33,003ms] [Warning] [omni.usd] Coding Error: in _ValidateEditPrim at line 1279 of C:\b\w\ca6c508eae419cf8\USD\pxr\usd\usd\stage.cpp -- Cannot create prim spec at path </__Master_4/robot0_C_forearm>; authoring to an instancing master is not allowed.

2022-10-13 01:01:44 [33,003ms] [Warning] [omni.usd] Warning: in AddAppliedSchema at line 300 of C:\b\w\ca6c508eae419cf8\USD\pxr\usd\usd\prim.cpp -- Unable to create primSpec at path </__Master_4/robot0_C_forearm> in edit target 'anon:00000189A4CFC0E0:World0.usd'. Failed to add applied API schema.

... (above warnings repeat)

This problem didn't occur when I run other tasks such as Humanoid and Cartpole.

My system spec: Windows 10 + AMD Ryzen 9 5900HX + RTX3080 laptop

Any help will be greatly appreciated!

Cant install on windows 10

PYTHON_PATH -m pip install -e .
this command doesnt work and it gives the following : The system cannot find the path specified.
even that I can check it with doskey /m and it gives me the correct path "doskey PYTHON_PATH=C:\Users\user\AppData\Local\ov\pkg\isaac_sim-\python.bat $" where I changed user to the correct path name on my PC

Setting num_envs=1 gives poor results when testing.

Hi,
Thank you for your contribution.

I tested the Anymal environment and all parameters were left unchanged.

After training the reward reaches above 65, but when I test with the environment number set to 1, the reward does not go above 10 and the robot can not walk properly.

However, when num_envs > 1 everything is fine on testing. What leads to this problem, and does this cause danger to the real robot, even though the simulation is already fine?

Getting LIDAR data in headless mode

How do I receive data in headless mode? It works when headless = False

change solver_velocity_iteration_count (and position) during simulation

Hello,

Is there a way to randomize these parameters DURING SIMULATION ? I have tried to put this code in the pre_physics step but there is no effect on the simulation (even tho the prameter of the simulation is well setted, i have checked with the getter). Is this some jit magic ? is there a way around it ?

        cfg = self._sim_config.parse_actor_config("myprim")
        cfg["solver_position_iteration_count"] = 4 # my new values
        cfg["solver_velocity_iteration_count"] = 1 # my new values
        self._sim_config.apply_articulation_settings("myprim", get_prim_at_path(self.aru.prim_path), cfg)

Regards,

Error when changing the number of environments

Hi,

Thank you for your work！

Runs with an error when I make changes to the number of environments, whether the number of environments is larger or less than the default number, for example:

PYTHON_PATH scripts/rlgames_train.py task=Cartpole num_envs=64

I got this:

Error executing job with overrides: ['task=Cartpole', 'num_envs=64']
Traceback (most recent call last):
File "scripts/rlgames_train.py", line 109, in parse_hydra_configs
rlg_trainer.run()
File "scripts/rlgames_train.py", line 82, in run
'sigma': None
File "/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/python/lib/python3.7/site-packages/rl_games/torch_runner.py", line 108, in run
self.run_train(args)
File "/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/python/lib/python3.7/site-packages/rl_games/torch_runner.py", line 86, in run_train
agent = self.algo_factory.create(self.algo_name, base_name='run', params=self.params)
File "/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/python/lib/python3.7/site-packages/rl_games/common/object_factory.py", line 15, in create
return builder(**kwargs)
File "/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/python/lib/python3.7/site-packages/rl_games/torch_runner.py", line 37, in
self.algo_factory.register_builder('a2c_continuous', lambda **kwargs : a2c_continuous.A2CAgent(**kwargs))
File "/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/python/lib/python3.7/site-packages/rl_games/algos_torch/a2c_continuous.py", line 18, in init
a2c_common.ContinuousA2CBase.init(self, base_name, params)
File "/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/python/lib/python3.7/site-packages/rl_games/common/a2c_common.py", line 965, in init
A2CBase.init(self, base_name, params)
File "/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/kit/python/lib/python3.7/site-packages/rl_games/common/a2c_common.py", line 185, in init
assert(self.batch_size % self.minibatch_size == 0)
AssertionError

Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
/home/jrl/.local/share/ov/pkg/isaac_sim-2022.1.0/python.sh: line 46: 8605 Segmentation fault (core dumped) $python_exe "$@" $args

Since I set a smaller number of environments, it seems that the problem is not related to memory.
Can you give me some advice?