Prior Policy Guided Dual-Agent Coordinated Manipulation Planning of Spacecraft-Manipulator System
2026-05-25 • Robotics
Robotics
AI summaryⓘ
The authors address the challenge of controlling a robot arm attached to a spacecraft without losing the spacecraft's stability. They created a system called DACMP that helps the robot arm move precisely while keeping the spacecraft steady. To train this system efficiently, they use a learning method guided by expert examples at specific times. Their tests show this method works better than existing ones, even in tough conditions like system limits and noises. They also made their code publicly available for others to try.
space manipulatorspacecraft attitudedynamic coupling6-DoF robotic armdeep reinforcement learningpolicy guidancetask success ratecontrol precisionenvironmental disturbancesperception uncertainties
Authors
Yuhui Hu, Dong Zhou, Kaihong Ouyang, Zhongliang Yu, Jianfeng Lv, Xiangyu Shao
Abstract
The strong dynamic coupling between the manipulator and the base poses a significant challenge to maintaining spacecraft attitude stability, potentially compromising mission safety. In this paper, we propose a Dual-Agent Coordinated Manipulation Planning (DACMP) framework that simultaneously achieves high-precision end-effector pose reaching for a 6-DoF space manipulator and attitude stabilization of the base spacecraft. To enhance learning efficiency, we present a prior policy-guided Deep Reinforcement Learning algorithm incorporating the Timestep-level Expert Switching Guidance (TESG) mechanism, thereby promoting global convergence and improving task success rates. Extensive experiments demonstrate that DACMP significantly outperforms baseline DRL algorithms in terms of task success rate and control precision. Furthermore, the robustness of DACMP is validated under various challenging scenarios, including system constraints, environmental disturbances, and perception uncertainties. The code and simulation configurations are available on GitHub: https://github.com/HIT-YuhuiHu/DACMP.