Project Author: Jie Yin at ([email protected])
- 2024.1.29 Accepted by ICRA2024. I have already released the datasets, and will release the code when I attend the ICRA in May,2024.
2. Next, I'm working on developing the ROS2 version of this project. If you are interested to collaborate to develop more functions, feel free to contact ([email protected]).
We introduce Ground-Fusion, a low-cost sensor fusion simultaneous localization and mapping (SLAM) system for ground vehicles. Our system features efficient initialization, effective sensor anomaly detection and handling, real-time dense color mapping, and robust localization in diverse environments. We tightly integrate RGB-D images, inertial measurements, wheel odometer and GNSS signals within a factor graph to achieve accurate and reliable localization both indoors and outdoors. To ensure successful initialization, we propose an efficient strategy that comprises three different methods: stationary, visual, and dynamic, tailored to handle diverse cases. Furthermore, we develop mechanisms to detect sensor anomalies and degradation, handling them adeptly to maintain system accuracy.
The preprint version of paper is arxiv. The dataset is at https://github.com/sjtuyinjie/M2DGR-plus and https://github.com/SJTU-ViSYS/M2DGR. And the homepage is at https://sites.google.com/view/ground-fusion/home.
Figure 1. We categorize corner cases into three types: visual, wheel, and GNSS challenges.
Tested on Ubuntu 18.04 (with ROS Melodic and OpenCV3) and on Ubuntu 20.04(with ROS Noetic and OpenCV4).
This package requires OpenCV 3/4 and some features of C++11.
This package requires Eigen 3.3.7, Ceres 1.14,Sophus and PCL 1.11.
This package also requires gnss_comm for ROS message definitions and some utility functions.
mkdir -p ~/Groundfusion_ws/src
cd ~/Groundfusion_ws/src
git clone https://github.com/HKUST-Aerial-Robotics/gnss_comm
git clone https://github.com/SJTU-ViSYS/Ground-Fusion
cd Ground-Fusion/thirdparty
cd ../..
catkin_make -j12
download at Ground-challenge and give a kind star.
# [launch] open a terminal and type:
source devel/setup.bash
roslaunch vins groundfusion.launch\
# [run localization] open another terminal:
source devel/setup.bash
rosrun vins vins_node src/Ground-Fusion/config/realsense/groundchallenge.yaml
# [dense map]open third terminal:
source devel/setup.bash
rosrun dense_map dense_map_node src/Ground-Fusion/config/realsense/groundchallenge.yaml
download at M2DGR-P and give a kind star.
# [launch] open a terminal and type:
source devel/setup.bash
roslaunch vins groundfusion.launch\
# [run localization] open another terminal:
source devel/setup.bash
rosrun vins vins_node src/Ground-Fusion/config/realsense/m2dgrp.yaml
# [dense map]open third terminal:
source devel/setup.bash
rosrun dense_map dense_map_node src/Ground-Fusion/config/realsense/m2dgrp.yaml
Thanks support from National Key R&D Program (2022YFB3903802), NSFC(62073214), and Midea Group. This project is based on GVINS, and has borrowed some codes from open-source projects VIW-Fusion and VINS-RGBD, thanks for your great contribution!
The source code of Ground-Fusion is released under GPLv3 license. Do not use this project for any commercial purpose unless permitted by authors. Yin Jie is still working on improving the system. For any technical issues, please contact him at [email protected].
If you use this work in an academic work, please cite:
@article{yin2021m2dgr,
title={M2dgr: A multi-sensor and multi-scenario slam dataset for ground robots},
author={Yin, Jie and Li, Ang and Li, Tao and Yu, Wenxian and Zou, Danping},
journal={IEEE Robotics and Automation Letters},
volume={7},
number={2},
pages={2266--2273},
year={2021},
publisher={IEEE}
}
@article{yin2024ground,
title={Ground-Fusion: A Low-cost Ground SLAM System Robust to Corner Cases},
author={Yin, Jie and Li, Ang and Xi, Wei and Yu, Wenxian and Zou, Danping},
journal={arXiv preprint arXiv:2402.14308},
year={2024}
}