A simple PointCloud Pipeline for ground based robots.
- Passthrough Filter - Cropping PointClouds to a range of acceptable and usable values.
- Voxel Filter - Downsampling PointClouds to speed up more computationally intensive filters.
- Segementation - Separating parts of PointClouds to ease in map building. Examples include: Ground Points, Obstacle Points, Overhead Points
- Post-Processor - Using segemented PointCloud indicies to generate new fully filtered PointClouds, or doing a 2D projection for scan based mapping.
- PointCloud-Concatenate - Merge multiple sources of Pointclouds into one Pointcloud.
This can be used to fix strange ign_gazebo frame_ids which come from a URDF robot model being spawned in to the simulator with an attached gpu_ray.