Skip to content

Latest commit

 

History

History
137 lines (100 loc) · 4.66 KB

README.md

File metadata and controls

137 lines (100 loc) · 4.66 KB
Raw

LeGO-LOAM-SR

This code is a fork from LeGO-LOAM-SR to migrate LeGO-LOAM algorithm to ROS2 Humble. Some syntax difference for ROS Humble is modified.

This code does not modify and/or improve the original LeGO-LOAM algorithm.

1. About the original LeGO-LOAM

LeGO-LOAM contains code for a lightweight and ground optimized lidar odometry and mapping (LeGO-LOAM) system for ROS compatible UGVs. The system takes in point cloud from a Velodyne VLP-16 Lidar (palced horizontal) and it outputs 6D pose estimation in real-time. A demonstration of the system can be found here -> https://www.youtube.com/watch?v=O3tz_ftHV48

drawing

2. Dependencies

  • ROS2 (tested with humble)
  • gtsam (Georgia Tech Smoothing and Mapping library) 
      git clone https://github.com/borglab/gtsam
  cd gtsam && git checkout 4.2a9
  mkdir build && cd build
  # For Ubuntu 22.04, add -DGTSAM_USE_SYSTEM_EIGEN=ON
  cmake .. -DGTSAM_BUILD_EXAMPLES_ALWAYS=OFF \
           -DGTSAM_BUILD_TESTS=OFF \
           -DGTSAM_WITH_TBB=OFF \
           -DGTSAM_BUILD_WITH_MARCH_NATIVE=OFF
  make -j$(nproc)
  sudo make install

3. Compile

You can use the following commands to download and compile the package.

cd ~/dev_ws/src
git clone https://github.com/theinge9/LeGO-LOAM-SR.git
cd ..
colcon build

4. The system

LeGO-LOAM is speficifally optimized for a horizontally placed lidar on a ground vehicle. It assumes there is always a ground plane in the scan.

The package performs segmentation before feature extraction.

drawing

Lidar odometry performs two-step Levenberg Marquardt optimization to get 6D transformation.

drawing

5. New sensor and configuration

To customize the behavior of the algorithm or to use a lidar different from VLP-16, edit the file config/loam_config.yaml.

One important thing to keep in mind is that our current implementation for range image projection is only suitable for sensors that have evenly distributed channels. If you want to use our algorithm with Velodyne VLP-32c or HDL-64e, you need to write your own implementation for such projection.

If the point cloud is not projected properly, you will lose many points and performance.

The IMU has been remove from the original LeGO-LOAM code.

6. Run the package

System can be started using the following command:

ros2 launch lego_loam_sr run.launch.py

Some sample bags can be downloaded from here.

6.1 Run ROS1 rosbag

To use rosbags from ROS1, you need to use a bridge between ROS1 and ROS2.

Play rosbag using ROS1:

source /opt/ros/melodic/setup.bash
roscore &
rosbag play *.bag --clock --topic /velodyne_points

Create the bridge in another terminal:

source /opt/ros/melodic/setup.bash
source /opt/ros/dashing/setup.bash
export ROS_MASTER_URI=http://localhost:11311
ros2 run ros1_bridge dynamic_bridge

Finally, run ROS2 code:

source /opt/ros/dashing/setup.bash
source ~/dev_ws/install/setup.bash
ros2 launch lego_loam_sr run.launch.py

6.2 New data-set

This dataset, Stevens data-set, is captured using a Velodyne VLP-16, which is mounted on an UGV - Clearpath Jackal, on Stevens Institute of Technology campus. The VLP-16 rotation rate is set to 10Hz. This data-set features over 20K scans and many loop-closures.

drawing

drawing

7. Cite LeGO-LOAM

Thank you for citing our LeGO-LOAM paper if you use any of this code:

@inproceedings{legoloam2018,
  title={LeGO-LOAM: Lightweight and Ground-Optimized Lidar Odometry and Mapping on Variable Terrain},
  author={Tixiao Shan and Brendan Englot},
  booktitle={IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages={4758-4765},
  year={2018},
  organization={IEEE}
}