中文摘要：

针对测量设备在多空间机器人在轨服务系统中的配置特点，提出了基于分布式滤波的非合作目标联合定位方法．首先定义了基于多空间机器人的非合作目标联合定位任务，随后建立了联合相对测量模型，推导了多空间机器人相对动力学的一般形式；设计了集中式联合定位滤波算法；在此基础上，为减少滤波迭代过程中的计算量，进一步提出了基于分布式滤波的联合定位算法；采用线性系统能观性理论，对滤波算法的稳定性进行了评估．最后，搭建了半物理实时仿真系统，对基于分布式滤波的联合定位算法进行了仿真验证，仿真结果表明联合定位算法具有较好的稳定性和较高的精度．

英文摘要：

According to the characteristics of relative measurement elements in the on-orbit servicing system with multiple space robots, a non-cooperative target co-localization strategy based on distributed filters is proposed. Firstly, the mission of non-cooperative target co-localization based on multiple space robots is defined, and a co-measurement model is also established. After that, a general expression of relative dynamics of multiple space robots is derived, and a centralized co-localization filter is designed. In order to reduce the computation load of iteration process in filtering, a co-localization method based on distributed filters is proposed. The stabilities of both filters are evaluated with linear system observability theory. Finally, a real-time semi-physical simulation system is established, and the co-location strategy based on distributed filters is simulated. The high stability and accuracy of the co-localization strategy are demonstrated by simulations.