https://github.com/PengYu-Team/Co-LRIO
A ROS2 package of CoLRIO: LiDAR-Ranging-Inertial Centralized State Estimation for Robotic Swarms.
https://github.com/PengYu-Team/Co-LRIO
centralized collaborative-slam lidar-ranging-inertial
Last synced: over 1 year ago
JSON representation
A ROS2 package of CoLRIO: LiDAR-Ranging-Inertial Centralized State Estimation for Robotic Swarms.
- Host: GitHub
- URL: https://github.com/PengYu-Team/Co-LRIO
- Owner: PengYu-Team
- License: apache-2.0
- Created: 2023-09-16T05:47:49.000Z (almost 3 years ago)
- Default Branch: master
- Last Pushed: 2024-02-29T07:50:16.000Z (over 2 years ago)
- Last Synced: 2024-10-27T19:00:47.638Z (over 1 year ago)
- Topics: centralized, collaborative-slam, lidar-ranging-inertial
- Language: C++
- Homepage:
- Size: 353 KB
- Stars: 80
- Watchers: 2
- Forks: 7
- Open Issues: 6
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# CoLRIO
A ROS2 package of CoLRIO: LiDAR-Ranging-Inertial Centralized State Estimation for Robotic Swarms.
https://github.com/PengYu-Team/Co-LRIO/assets/41199568/81985d82-983c-4eca-898b-43e8f84e7b45
## Prerequisites
- [Ubuntu ROS2 Foxy](http://wiki.ros.org/ROS/Installation) (Robot Operating System 2 on Ubuntu 20.04)
- CMake (Compilation Configuration Tool)
- [PCL](https://pointclouds.org/downloads/linux.html) (Default Point Cloud Library on Ubuntu work normally)
- [Eigen](http://eigen.tuxfamily.org/index.php?title=Main_Page) (Default Eigen library on Ubuntu work normally)
- [GTSAM 4.2a8](https://github.com/borglab/gtsam/releases) (Georgia Tech Smoothing and Mapping library)
## Compilation
Build CoLRIO:
```
mkdir -p ~/cslam_ws/src
cd ~/cslam_ws/src
git clone https://github.com/PengYu-Team/Co-LRIO.git
cd ../
colcon build --symlink-install
```
## Run with Dataset
- [our dataset] TBD.
- [S3E dataset](https://github.com/PengYu-Team/S3E). The datasets are configured to run with default parameter.
```
ros2 launch co_lrio run.launch.py
ros2 bag play *your-bag-path*
```
## Citation
This work is published in IEEE ICRA 2024 conference, and please cite related papers:
```
@misc{zhong2024colrio,
title={CoLRIO: LiDAR-Ranging-Inertial Centralized State Estimation for Robotic Swarms},
author={Shipeng Zhong and Hongbo Chen and Yuhua Qi and Dapeng Feng and Zhiqiang Chen and Jin Wu and Weisong Wen and Ming Liu},
year={2024},
eprint={2402.11790},
archivePrefix={arXiv},
primaryClass={cs.RO}
}
```
```
@article{feng2022s3e,
title={S3e: A large-scale multimodal dataset for collaborative slam},
author={Feng, Dapeng and Qi, Yuhua and Zhong, Shipeng and Chen, Zhiqiang and Jiao, Yudu and Chen, Qiming and Jiang, Tao and Chen, Hongbo},
journal={arXiv preprint arXiv:2210.13723},
year={2022}
}
```
## Acknowledgement
- We combined the front end of CoLRIO and the [DLO](https://github.com/vectr-ucla/direct_lidar_odometry) to achieve the 5th position in the [ICCV 2023 LiDAR-Inertial SLAM Challenge](https://superodometry.com/iccv23_challenge_LiI).
The Leaderboard is shown as follow:
And the hardware and results are shown as follow:
- CoLRIO depends on [FAST-GICP](https://github.com/SMRT-AIST/fast_gicp) (Kenji Koide, Masashi Yokozuka, Shuji Oishi, and Atsuhiko Banno, "Voxelized GICP for fast and accurate 3D point cloud registration".).
- CoLRIO depends on [GncOptimizer](https://github.com/borglab/gtsam/blob/3a1fe574683f608759eaff4636ab53def600ce84/gtsam/nonlinear/GncOptimizer.h#L45) (Yang, Antonante, Tzoumas, Carlone, "Graduated Non-Convexity for Robust Spatial Perception: From Non-Minimal Solvers to Global Outlier Rejection").