Skip to content
Forestry-Robotics-UCPublic

About

No description, website, or topics provided.

Resources

Stars

1 star

Watchers

0 watching

Forks

Latest commit

 

History

31 Commits

Folders and files

Repository files navigation

FRUC iG-LIO

Preserves the functionality of the original code, but introduces some important modifications:

  • Removes dependencies (and support) of Livox LiDARs;
  • Adds support to the legacy Velodyne Velarray M1600 LiDAR (by Pedro Tomás);
  • Fixes the point cloud undistorting mechanism (at least with the Ouster) by fixing Ouster-related code to update old message field names, types and other small details;
  • Adds a well-tuned configuration file to use with the Ouster OS1 Rev7 128 channel LiDAR + internal IMU;

Acknowledgement:

The original repository can be found here.


To Install Dependencies

cd ~/catkin_ws && rosdep install --from-paths src --ignore-src -r -y

❗ Note: Change ~/catkin_ws to your actual workspace path before running the command.

Instructions to Use

Let's consider, as an example, that we are using the Ouster OS1 Rev7 128 channel LiDAR and internal IMU. Then, if we want to run iG-LIO online, we simple run roslaunch ig_lio lio_ouster_128_online.launch with the LiDAR driver running in parallel and iG-LIO automatically starts publishing the map -> base_link transform on the topic /tf.

If, instead we want to run it offline, we run roslaunch ig_lio lio_ouster_128.launch (or simple turn the use_sim_time parameter to true).

To use with different LiDAR/IMU combos, create a new config file or modify an existing one to adapt it to your needs, keeping in mind that the translation vector and rotation matrix provided must map the points from the LiDAR frame to the IMU frame, not the other way around!


Tested in docker with osrf:ros-noetic-full-focal.

About

No description, website, or topics provided.

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages