利用空间机器人对大惯量非合作目标开展在轨维护是合理使用轨道资源、提升航天器经济效益的重要手段之一。大惯量非合作目标处于翻滚状态、初始动量大且惯性参数未知，这为捕获任务带来了新的挑战。本项目针对大惯量非合作目标捕获后多臂空间机器人难以协同控制的问题展开研究。首先，提出一种动力学耦合因子的概念，并结合耦合椭圆量化分析空间机器人与目标之间的耦合程度；其次，根据耦合程度，利用可操作度的概念计算多臂空间机器人的最大可操作度，确定系统最佳臂型；之后，利用广义相对雅可比矩阵进行多臂协同运动规划，保证卫星基座姿态扰动最小；最后，通过动量守恒辨识目标惯性参数，结合自适应反作用零空间算法实现在参数辨识过程中的基座姿态控制。本项目的意义是根据我国空间机器人发展的现实需求，探索新的多臂空间机器人在轨捕获大惯量航天器的控制思路，以求在关键技术上有所突破和创新，推动空间机器人技术和相关学科的发展。

Nowadays, the space robots play an important role in space exploration. This research project focuses on the task of capturing a large non-cooperative target with multi-arm space robot. When the multi-arm space robot captures a large non-cooperative target, the dynamic coupling between the base and manipulators together with the target are more complex. Dynamic model for multi-arm space robotic system will be developed. Firstly, the dynamic coupling factors and coupling ellipsoids are proposed as two tools to measure the coupling degree. Secondly, by using the concept of manipulability measure, a performance index is introduced and used to design and choose an optimization configuration for a multi-arm space robot. Then, the generalized relative Jacobin matrix (GRJM) is defined based on the kinematic and momentum conservation equations of space robots for multi-arm coordinated operation in space. Lastly, to identify the minimal inertial parameters of the captured target, a new momentum model-based method is proposed, which is critical to motion planning and control of space robots. From the development trends in on-orbit servicing, multi-arm space robot will be one of the most promising approaches in future. This project mainly aims at providing a reference for the practical application of space robot technology and making scientific breakthroughs in this field.