A Surprisingly Efficient Representation for Multi-Finger Grasping

TL;DR

This paper introduces an efficient representation for multi-finger grasping that enables high success rates with minimal training data, facilitating real-world robotic grasping and handover tasks.

Contribution

It proposes a novel grasp representation and a simple decision model that significantly reduces training data requirements for effective multi-finger grasping.

Findings

Achieves 78.64% success with only 500 training samples

Reaches 87% success with 4500 samples in real-world tests

Attains 84.51% success in dynamic human-robot handover scenarios

Abstract

The problem of grasping objects using a multi-finger hand has received significant attention in recent years. However, it remains challenging to handle a large number of unfamiliar objects in real and cluttered environments. In this work, we propose a representation that can be effectively mapped to the multi-finger grasp space. Based on this representation, we develop a simple decision model that generates accurate grasp quality scores for different multi-finger grasp poses using only hundreds to thousands of training samples. We demonstrate that our representation performs well on a real robot and achieves a success rate of 78.64% after training with only 500 real-world grasp attempts and 87% with 4500 grasp attempts. Additionally, we achieve a success rate of 84.51% in a dynamic human-robot handover scenario using a multi-finger hand.