Tabletop Object Perception
This page describes the sensor data processing tools that allow complete usage of the object manipulation functionality contained in object_manipulation and pr2_object_manipulation. However, it is not a development goal in itself. Other (better) implementations of sensor processing for manipulation are under active development. This framework is meant to allow using the manipulation functionality until other alternatives mature.
The main relevant packages are:
tabletop_object_detector: provides object segmentation and recognition services, for constrained scenes.
tabletop_collision_map_processing: adds the output of the object detector to the collision environment, provides detected objects with collision names, oversees the process of taking static collision maps, etc.
Running the Manipulation Pipeline
To launch the manipulation pipeline, complete with the sensor processing provided here, and execute pickup and place tasks using the PR2 robot, tutorials and launch files are provided in pr2_tabletop_manipulation_apps.
Report a Bug