This repository provides the implementation for PERCH.

The package 'kinect_sim' is based on PCL's simulation API and serves as the depth-image renderer/simulator for PERCH. This provides the functionality to generate the depth-image for a scene, given a camera pose, set of 3D models/meshes and their poses.

The 'sbpl_perception' package implements the PERCH algorithm. This works in conjunction with the SBPL planning library (https://github.com/sbpl/sbpl) and provides the 'search environment'. The documentation in the SBPL github page provides examples of the planner-environment usage.

• ROS Hydro+
• C++11
• OpenCV 2.x
• PCL 1.7+

1. Get ROS Hydro (or Indigo) from http://wiki.ros.org/hydro/Installation/Ubuntu
2. Set up a catkin workspace ~/my_workspace (http://wiki.ros.org/catkin/Tutorials/create_a_workspace).
3. Download the rosinstall file to your workspace.

cd ~/my_workspace
wstool init src
wstool merge -t src perch.rosinstall
wstool update -t src
catkin_make

An example scene is provided under sbpl_perception/demo. To run PERCH,

Example usage:

roscd sbpl_perception && mkdir visualization
roslaunch sbpl_perception demo.launch 

The states 'expanded' as well as the goal state will be saved under sbpl_perception/visualization. The expanded states will also show up on an OpenCV window named "Expansions". To also save all the 'generated' (rendered) states to sbpl_perception/visualization, use

roslaunch sbpl_perception demo.launch image_debug:=true

(TODO: more details)

Run wstool update -t src to pull changes from all related repositories.