iROSA-lab/reachability_map_visualizer

Visualizer for robot reachability and base placement maps (based on reuleaux)

Reachability map visualizer

ROS package for visualizing a robot's reachability and base placement maps

• Currently supports ROS melodic and Ubuntu 18 (although porting to ROS noetic and Ubuntu 20 should not be difficult).

• To generate the reachability maps, have a look at the sampled_reachability_maps (github.com/iROSA-lab/sampled_reachability_maps) repository.

• This repository is a part of the codebase of the paper: Robot Learning of Mobile Manipulation With Reachability Behavior Priors [1] [Paper] [Project site]

• The code is based on the workspace_visualization package of the reuleaux repository (github.com/ros-industrial-consortium/reuleaux). Please refer to wiki.ros.org/reuleaux for more information.

Loading a map and running the visualization

• Build this package in your ROS workspace and source the devel/setup.bash file

• Load the reachability/base_placement map by running (for example)

  rosrun reachability_map_visualizer load_reachability_map $(rospack find reachability_map_visualizer)/maps/3D_reachability_map_arm_left_tool_link_0.05_2022.h5
  

• Run rviz and in the 'Displays' panel click on 'Add'. Choose 'By display type' and select the ReachabilityMap option under the 'reachability_map_visualizer' folder. To run our example rviz config, use:

  rosrun rviz rviz -d $(rospack find reachability_map_visualizer)/rviz/reachability_map.rviz
  

  

• You can play with the visualization settings in Rviz:

  

• The voxel coloring scheme can be changed by editing the "ReachMapVisual::setColorSpherebyRI" function in the "reachability_map_visual "file. By default, dark blue denotes the highest reachability while red denotes the lowest.

References

[1] S. Jauhri, J. Peters and G. Chalvatzaki, "Robot Learning of Mobile Manipulation With Reachability Behavior Priors", https://doi.org/10.1109/LRA.2022.3188109