Dexterous Manipulation Based on Prior Dexterous Grasp Pose Knowledge

TL;DR

This paper presents a reinforcement learning method for dexterous manipulation that uses prior grasp pose knowledge to improve efficiency and success rates, by decoupling grasp generation and manipulation exploration.

Contribution

It introduces a novel two-phase approach that leverages prior grasp knowledge, contrasting with fixed grasp pose methods, to enhance manipulation performance.

Findings

Significant improvements in learning efficiency

Higher success rates across four tasks

Most learning time spent on initial positioning and viewpoint selection

Abstract

Dexterous manipulation has received considerable attention in recent research. Predominantly, existing studies have concentrated on reinforcement learning methods to address the substantial degrees of freedom in hand movements. Nonetheless, these methods typically suffer from low efficiency and accuracy. In this work, we introduce a novel reinforcement learning approach that leverages prior dexterous grasp pose knowledge to enhance both efficiency and accuracy. Unlike previous work, they always make the robotic hand go with a fixed dexterous grasp pose, We decouple the manipulation process into two distinct phases: initially, we generate a dexterous grasp pose targeting the functional part of the object; after that, we employ reinforcement learning to comprehensively explore the environment. Our findings suggest that the majority of learning time is expended in identifying the appropriate initial position and selecting the optimal manipulation viewpoint. Experimental results demonstrate significant improvements in learning efficiency and success rates across four distinct tasks.