AnyDexGrasp: General Dexterous Grasping for Different Hands with Human-level Learning Efficiency

TL;DR

This paper presents a human-level data-efficient approach for dexterous robotic grasping across various hands, using a two-stage model that generalizes scene understanding and hand-specific grasp execution, achieving high success rates with minimal training data.

Contribution

The method introduces a universal scene-to-grasp model combined with hand-specific training, significantly reducing data requirements compared to traditional approaches.

Findings

Achieves 75-95% grasp success rate across different robotic hands.

Requires only hundreds of grasp attempts for training.

Generalizes well to novel objects in cluttered environments.

Abstract

We introduce an efficient approach for learning dexterous grasping with minimal data, advancing robotic manipulation capabilities across different robotic hands. Unlike traditional methods that require millions of grasp labels for each robotic hand, our method achieves high performance with human-level learning efficiency: only hundreds of grasp attempts on 40 training objects. The approach separates the grasping process into two stages: first, a universal model maps scene geometry to intermediate contact-centric grasp representations, independent of specific robotic hands. Next, a unique grasp decision model is trained for each robotic hand through real-world trial and error, translating these representations into final grasp poses. Our results show a grasp success rate of 75-95\% across three different robotic hands in real-world cluttered environments with over 150 novel objects, improving to 80-98\% with increased training objects. This adaptable method demonstrates promising applications for humanoid robots, prosthetics, and other domains requiring robust, versatile robotic manipulation.