armlabcuicar / nsf-hc10dt

This repository contains the gazebo model and MoveIt! plugins for the Motoman HC10-DT with a Velodyne VLP-16 lidar as an end-effector for 3D mapping.

Commands

Gazebo roslaunch motoman_hc10_gazebo hc10dt_gazebo.launch

MoveIt! roslaunch motoman_hc10dt_moveit_config moveit_planning_execution.launch

python rosrun move_hc0dt trial.py

Dependencies

Ensure that the following package dependencies are installed in your system:

• ros-kinetic-ros-control
• ros-kinetic-ros-controllers
• ros-kinetic-controller-interface
• ros-kinetic-controller-manager
• ros-kinetic-controller-manager-msgs
• ros-kinetic-gazebo-ros
• ros-kinetic-gazebo-ros-control
• ros-kinetic-joint-state-controller
• ros-kinetic-joint-trajectory-controller
• ros-kinetic-effort-controllers
• ros-kinetic-velocity-controllers

Known Errors

[ERROR] [1589821347.035206609, 0.367000000]: No p gain specified for pid.  Namespace: /gazebo_ros_control/pid_gains/joint_1_s

This is not really an error. The PID parameters are optional if you are using gazebo_ros_control. The error comes because when initializing RobotHWSim, the robot_ros_control tries to load the pid_gains. Basically, if pid parameters were found, gazebo_ros_control will use pid controllers in ROS to control the joints (velocity or position) by effort. Otherwise, the joints will be controlled with gazebo methods.

https://answers.ros.org/question/293830/what-is-the-fix-for-no-p-gain-specified-for-pid-namespace-gazebo_ros_controlpid_gainsback_right_wheel_joint-ros-melodic/

Git Workflow

Forking

Create a fork of armlab-clemson/nsf-hc10dt in your own account. Clone this fork into your computer.

$ mkdir ~/git && cd ~/git
$ git clone --recurse-submodules https://github.com/USER/nsf-hc10dt.git
$ cd nsf-hc10dt

Notice that the option --recurse-submodules is passed to git-clone. This automatically initializes and updates any included submodules.

The ros-stacks directory may contains external repositories as Git submodules. One example is the motoman_experimental repository. If the directory is empty (perhaps you forgot --recurse-submodules when cloning), you'll have to perform the following to get the code for each submodule.

~/git/arm-scr$ cd ros-stacks/motoman_experimental
$ git submodule init
$ git submodule update 

Adding remotes

Next, set up additional remotes for this repository. The one you cloned from, your fork, will be named origin by default. Then, we'll refer to the mainline respository as the upstream repository.

~/git/nsf-hc10dt$ git remote -v # inspect remotes
origin	https://github.com/USER/nsf-hc10dt.git (fetch)
origin	https://github.com/USER/nsf-hc10dt.git (push)
$ git remote add upstream https://github.com/armlab-clemson/nsf-hc10dt.git
$ git remote -v
origin	https://github.com/USER/nsf-hc10dt.git (fetch)
origin	https://github.com/USER/nsf-hc10dt.git (push)
upstream	https://github.com/armlab-clemson/nsf-hc10dt.git (fetch)
upstream	https://github.com/armlab-clemson/nsf-hc10dt.git (push)

Branches

Next you should look at your branches and create a new one for your next development topic so you can start making changes safely. Suppose you're working on a fix to the colors used in the OTTO1500 model.

~/git/nsf-hc10dt$ git branch # inspect branches, master is current
* master

$ git branch dev-1.1 #create branch fix-otto-colors
$ git checkout dev-1.1 # check out branch fix-otto-colors
$ git branch # inspect branches, fix-otto-colors is current
* dev-1.1
  master

Now go make some changes to fix the colors, test them, add them, and commit them. Assuming that you only changed the XACRO macro for OTTO:

~/git/nsf-hc10dt$ git add ros-stacks/clearpath_otto1500_support/urdf/otto1500_macro.xacro
$ git commit -m 'fix otto colors'
$ git log # inspect the commit history for the current branch
commit xxxxxxx...
Author: User Name <[email protected]>
Date:   ...

fix otto colors

Repeat this edit/test/commit cycle until you're ready to share your changes. Then push your topic branch to your fork remember that the remote alias for your fork is origin.

~/git/nsf-hc10dt$ git push origin dev-1.1

Finally, after making all the changes, merge the branch with the master.

~/git/nsf-hc10dt$ git checkout master
~/git/nsf-hc10dt$ git merge dev-1.1

In the browser GUI, you will see a new pull request to merge the branches. Click on the pull request and accept it. Resolve the conflicts - if there are any - and merge the branches with comments if needed.

Now delete the branch after merging.

~/git/nsf-hc10dt$ git branch -d dev-1.1

Pull Requests

To pull your changes into the armlab-clemson/nsf-hc10dt repository, submit a pull request to the admin. Once you push the changes to your repo, Compare & pull request buttons will appear in GitHub. Select compare and after resolving conflicting changes, submit the pull request.

Git Workspace Configuration

nsf_hc10dt/
    ros-stacks/
	motoman/
        industrial_core/

ROS Workspace Configuration

ros_ws/
    build/
    devel/
    src/
	<symbolic links from ros-stacks>

Adding packages to the src directory in this workspace

We'll do this by going into src and creating symbolic links to the desired ROS packages or stacks that we've cloned from Git repositories.

~/ros_ws$ cd src

~/ros_ws/src$ ln -s ~/nsf-robot-project/nsf-hc10dt/ros-stacks/motoman

Then go up a level and build the workspace again to compile the package(s).

~/ros_ws/src$ cd ..
$ catkin_make

As you checkout different branches your symbolic links may become invalid. This is okay. Just build the workspace with bad links and those targets will be skipped.

Developers:

• Mark T
• Adhiti R

nsf-hc10dt's Issues

<gravity> tag in Gazebo Simulation getting overwritten to false

upon launching the gazebo command given in ReadMe.md , it throws an error that the gravity tag gets overwritten. Confirmed by placing the objects in the air, they don't fall.

Warning [parser_urdf.cc:1232] multiple inconsistent <gravity> exists due to fixed joint reduction overwriting previous value [true] with [false]. Warning [parser_urdf.cc:1232] multiple inconsistent <gravity> exists due to fixed joint reduction overwriting previous value [false] with [true].