6-DOF Grasping for Target-driven Object Manipulation in Clutter

TL;DR

This paper introduces a method for planning 6-DOF grasps in cluttered environments from partial observations, significantly improving success rates and enabling complex object retrieval sequences in real-world robotic manipulation.

Contribution

It presents a novel approach that enables 6-DOF grasp planning for any object in clutter, surpassing existing top-down methods and incorporating learned collision checking for effective manipulation sequences.

Findings

Achieved 80.3% grasp success rate, outperforming baselines by 17.6%.

Successfully cleared 9 cluttered scenes with 23 objects and 51 picks.

Demonstrated effective grasp sequencing for inaccessible objects.

Abstract

Grasping in cluttered environments is a fundamental but challenging robotic skill. It requires both reasoning about unseen object parts and potential collisions with the manipulator. Most existing data-driven approaches avoid this problem by limiting themselves to top-down planar grasps which is insufficient for many real-world scenarios and greatly limits possible grasps. We present a method that plans 6-DOF grasps for any desired object in a cluttered scene from partial point cloud observations. Our method achieves a grasp success of 80.3%, outperforming baseline approaches by 17.6% and clearing 9 cluttered table scenes (which contain 23 unknown objects and 51 picks in total) on a real robotic platform. By using our learned collision checking module, we can even reason about effective grasp sequences to retrieve objects that are not immediately accessible. Supplementary video can be found at https://youtu.be/w0B5S-gCsJk.