Refining 6-DoF Grasps with Context-Specific Classifiers

TL;DR

This paper introduces GraspFlow, a modular and simple method for refining 6-DoF grasps using context-specific classifiers, achieving high success rates on real-world robot experiments.

Contribution

It presents a novel discriminator gradient-flow approach for grasp refinement that is modular, easy to implement, and effective across diverse objects and criteria.

Findings

94% success rate on first grasp attempt across 20 objects

100% success rate by second grasp attempt

Functional grasp quality improved by up to 33% with handover discriminator

Abstract

In this work, we present GraspFlow, a refinement approach for generating context-specific grasps. We formulate the problem of grasp synthesis as a sampling problem: we seek to sample from a context-conditioned probability distribution of successful grasps. However, this target distribution is unknown. As a solution, we devise a discriminator gradient-flow method to evolve grasps obtained from a simpler distribution in a manner that mimics sampling from the desired target distribution. Unlike existing approaches, GraspFlow is modular, allowing grasps that satisfy multiple criteria to be obtained simply by incorporating the relevant discriminators. It is also simple to implement, requiring minimal code given existing auto-differentiation libraries and suitable discriminators. Experiments show that GraspFlow generates stable and executable grasps on a real-world Panda robot for a diverse range of objects. In particular, in 60 trials on 20 different household objects, the first attempted grasp was successful 94% of the time, and 100% grasp success was achieved by the second grasp. Moreover, incorporating a functional discriminator for robot-human handover improved the functional aspect of the grasp by up to 33%.