Pointcloud Merger

Installation

catkin build pointcloud_merger

Usage

Fusing pointclouds from multiple time stamps

roslaunch pointcloud_merger pointcloud_filter_examples.launch

If you are replaying a bag, do not forget to use sim time.

rosparam set /use_sim_time true

This file contains a filter chain to merge pointclouds temporally, crop and voxelize them spacially into a single point cloud.

Parameters for merging

• cloud_in: input point cloud
• cloud_out: output point cloud
• odom: odometry topic
• fixed_frame: odometry frame
• min_time_between_input: minimum time before new measurement is added [s]
• min_translation_between_input: minimum translation before new measurement is added [m]
• min_rotation_between_input: minimum rotation before new measurement is added [rad]
• transform_tolerance: allowable delay in TF data [s]

Parameters for cropping

• reference_frame: reference frame to crop in
• filter_field_name: axis to crop along
• filter_limit_min: min value for filter
• filter_limit_max: max value for filter

Parameters for voxelizing

• voxel_size: size of voxels (cube) to aggregate points [m]

Credits

Robin Schmid, schmidrobin@outlook.com