Learning to Efficiently Plan Robust Frictional Multi-Object Grasps
This addresses the efficiency of robotic decluttering tasks, offering incremental improvements over frictionless multi-object grasping.
The paper tackles the problem of decluttering multiple rigid convex polygonal objects by introducing friction into multi-object grasping to increase efficiency, resulting in a 13.7% higher success rate, 1.6x more picks per hour, and 6.3x faster planning compared to prior multi-object methods.
We consider a decluttering problem where multiple rigid convex polygonal objects rest in randomly placed positions and orientations on a planar surface and must be efficiently transported to a packing box using both single and multi-object grasps. Prior work considered frictionless multi-object grasping. In this paper, we introduce friction to increase the number of potential grasps for a given group of objects, and thus increase picks per hour. We train a neural network using real examples to plan robust multi-object grasps. In physical experiments, we find a 13.7% increase in success rate, a 1.6x increase in picks per hour, and a 6.3x decrease in grasp planning time compared to prior work on multi-object grasping. Compared to single-object grasping, we find a 3.1x increase in picks per hour.