FAST_livo_M2DGRP

FAST-LIVO

This file uses external coordinate conversion and lidar format conversion to convert the M2DGR-plus lidar and IMU to a coordinate system. Adds timestamps and ring information to the lidar data to transform /rslidar_points into /velodyne_points format.

1.Dependency

1.1 Ubuntu and ROS

Tested with Ubuntu 20.04 and ROS Noetic.ROS Installation.

1.2 Eigen

Tested with Eigen 3.3.7.Eigen Installation.

1.3 PCL

Tested with PCL 1.10 comes with ROS Noetic.PCL installation.

1.4 OpenCV

Tested with OpenCV 4.2.0 comes with ROS Noetic.OpenCV installation.

1.5 Sophus

Sophus Installation for the non-templated/double-only version.

git clone https://github.com/strasdat/Sophus.git
cd Sophus
git checkout a621ff
mkdir build && cd build && cmake ..
make
sudo make install

1.6 Livox SDK

Because need to install livox_ros_driver, install the livox SDK first.

git clone https://github.com/Livox-SDK/Livox-SDK.git
cd Livox-SDK
cd build && cmake ..
make
sudo make install

2. Compile

cd ~/catkin_ws/src
git clone https://github.com/sjtuyinjie/M2DGR-Benchmark.git && cd M2DGR-Benchmark && git sparse-checkout set --no-cone FAST_livo_M2DGRP
cd ../..
catkin_make

3. Run M2DGR example

source devel/setup.bash
roslaunch fast_livo mapping_m2dgr.launch
rosbag play door_02.bag 

4. Run M2DGR-plus example

source devel/setup.bash
roslaunch fast_livo mapping_m2dgr_plus.launch

//imu coordinate conversion
source devel/setup.bash
roslaunch imu_transformer ned_to_enu.launch

//lidar format conversion
source devel/setup.bash
rosrun rs_to_velodyne rs_to_velodyne m2dgrplus XYZIRT

rosbag play anomaly.bag --topic /rslidar_points /camera/color/image_raw /camera/imu