The tight integration of FAST-LIO with Radius-Search-based loop closure module.
FAST-LIO (Fast LiDAR-Inertial Odometry) is a computationally efficient and robust LiDAR-inertial odometry package. It fuses LiDAR feature points with IMU data using a tightly-coupled iterated extended Kalman filter. But it doesn't have a loop closure module to eliminate the accumulated drift.
Therefore, this project implements the pose graph optimization with a radius-search-based loop closure module which refers to FAST_LIO_SLAM. And the pose and map in the iterated extended Kalman filter of FAST-LIO will be updated according to the optimization which is a key difference with FAST_LIO_SLAM.
- Ubuntu 18.04 and ROS Melodic
- PCL >= 1.8 (default for Ubuntu 18.04)
- Eigen >= 3.3.4 (default for Ubuntu 18.04)
- GTSAM >= 4.0.0(tested on 4.0.0-alpha2)
cd YOUR_WORKSPACE/src
git clone https://github.com/yanliang-wang/FAST_LIO_LC.git
cd ..
catkin_make
You can test this project with our data which includes /velodyne_points
(10Hz) and /imu/data
(400Hz).
roslaunch fast_lio mapping_velodyne.launch
roslaunch aloam_velodyne fastlio_velodyne_VLP_16.launch
rosbag play T3F2-2021-08-02-15-00-12.bag -r 2
If you want to test the original FAST LIO (i.e. without the loop closure module), you can set
lc_enable
in themapping_velodyne.launch
tofalse
and run following commands.roslaunch fast_lio mapping_velodyne.launch rosbag play T3F2-2021-08-02-15-00-12.bag -r 2
video: Youtube link , Bilibili link
In this project, the LIO module refers to FAST-LIO and the pose graph optimization refers to FAST_LIO_SLAM.
Many thanks for their work.