空间机器人的同伦类约束路径规划  

Path Planning with Homotopy Class Constraints for Space Robots

作　　者：姚蔚然 田昊宇 张欧阳 曾祎 孙光辉 吴立刚[1] YAO Weiran;TIAN Haoyu;ZHANG Ouyang;ZENG Yi;SUN Guanghui;WU Ligang(School of Astronautics,Harbin Institute of Technology,Harbin 150001,Heilongjiang,China;National Key Laboratory of Complex System Control and Intelligent Agent Cooperation,Beijing 100074,China)

机构地区：[1]哈尔滨工业大学航天学院,黑龙江哈尔滨150001 [2]复杂系统控制与智能协同全国重点实验室,北京100074

出　　处：《上海航天（中英文）》2024年第3期63-73,共11页Aerospace Shanghai（Chinese&English）

基　　金：国家自然科学基金(U23A20346,62106062)。

摘　　要：同伦类约束路径规划是空间机器人路径规划的一种重要技术手段,能够有效实现复杂约束条件下路径规划的运算轻量化,具有重要的理论意义和应用价值。同伦路径规划以同伦这一数学工具作为基础,将空间机器人的可行路径集限制在范围有限且具有良好函数性质的同伦结构中,在该结构中进行优化路径的搜索,在保证路径规划有效性的同时降低了算法的复杂性。介绍了基于同伦的路径规划近似算法框架,以及路径同伦构造和规划问题的求解。从面向约束的路径同伦构造方法和同伦优化搜索2个层面分析了近年来针对空间机器人同伦路径规划研究的进展。根据研究现状,对空间机器人的同伦类约束路径规划的未来发展进行了展望。The path planning with homotopy class constraints is an important technical means for the path planning for space robots.It can effectively achieve lightweight computation of path planning under complex constraints,and has important theoretical significance and application value.In the path planning with homotopy class constraints,the set of robot′s feasible paths is limited within a homotopy structure with good functional properties.The search for optimized paths is carried out within the homotopy structure,which ensures the effectiveness of the path planning while reducing the complexity of the algorithm.In this paper,the approximation algorithm frameworks for homotopy-based path planning,the path homotopy construction methods,and the solutions to the planning problems are introduced.The progress of homotopy path planning for space robots in recent years is analyzed from two aspects,i.e.,path homotopy construction methods with constrains and homotopy optimization search.According to the current research situation,the future development of homotopy path planning methods for space robots is prospected.

关 键 词：空间机器人 路径规划 近似算法 同伦类 任务性能优化目标 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]