Reinforcement Learning with Prior Policy Guidance for Motion Planning of Dual-Arm Free-Floating Space Robot

TL;DR

This paper introduces EfficientLPT, a reinforcement learning algorithm that incorporates prior policy guidance and an improved reward function to enhance motion planning accuracy for dual-arm space robots, especially in capturing rotating objects.

Contribution

The paper presents a novel RL algorithm with a mixed policy and infinite norm-based reward, addressing pose constraints and dynamic coupling in dual-arm space robot planning.

Findings

Successfully captures rotating objects at different speeds

Improves planning accuracy with prior policy guidance

Handles pose constraints in complex space robot tasks

Abstract

Reinforcement learning methods as a promising technique have achieved superior results in the motion planning of free-floating space robots. However, due to the increase in planning dimension and the intensification of system dynamics coupling, the motion planning of dual-arm free-floating space robots remains an open challenge. In particular, the current study cannot handle the task of capturing a non-cooperative object due to the lack of the pose constraint of the end-effectors. To address the problem, we propose a novel algorithm, EfficientLPT, to facilitate RL-based methods to improve planning accuracy efficiently. Our core contributions are constructing a mixed policy with prior knowledge guidance and introducing infinite norm to build a more reasonable reward function. Furthermore, our method successfully captures a rotating object with different spinning speeds.