LIO-SAM Mapper

This is an offline version of LIO-SAM that produces a determenstic output, through exchanging messages/data using pointers instead of topics. It is configured to run with ouster lidar packets and pointclouds. This mapper can output different pointclouds to PCD files as well as a CSV file containing the poses at each keyframe.

Installation

Todo.

Demo

• Download and unzip court_yard_stroll_filtered.zip.
• Modify some parameters in param.yaml.
  • Set readBag to the path of court_yard_stroll_filtered.zip.
  • Set SaveDire to a directory where the output files could be saved.
  • Set pointCloudTopic to /os_cloud_node/points.
• Run

roslaunch lio_sam run.launch

Parameters

Below are the additional parameters added, information about the other parameters might be found in LIO-SAM.

Parameter	Type	Description
readBag	String	Rosbag path
saveDir	String	Saving directory
saveToRosbag	Bool	Set to True to save global map and trajectory to a rosbag
saveTrajectoryCSV	Bool	Set to True to save trajectory to CSV
saveRawPCD	Bool	Set to True to save raw lidar pointclouds to PCDc (lidar frame)
saveDeskewedPCD	Bool	Set to True to save deskewed pointclouds to PCDs (lidar frame)
saveRegisteredCloudPCD	Bool	Set to True to save registered pointclouds to PCDs (map frame)
saveRegisteredFeaturesPCD	Bool	Set to True to save registered feature pointclouds to PCDs (map frame)
LMOMaxIterations	Int	Max iterations count for lidar odometry optimisation