基于多目标粒子群优化算法的自由漂浮空间机器人负载最大化轨迹优化  

Load Maximization Trajectory Optimization for Free-Floating Space Robot Using Multi-objective Particle Swarm Optimization Algorithm

作　　者：刘勇[1] 贾庆轩[1] 陈钢[1] 孙汉旭[1] 

机构地区：[1]北京邮电大学自动化学院,北京100876

出　　处：《机器人》2014年第4期402-410,共9页Robot

基　　金：国家自然科学基金资助项目(61175080);国家973计划资助项目(2013CB733000);教育部高等学校博士学科点专项科研基金资助项目(20120005120004);中央高校基本科研业务费专项资金资助项目(2013PTB-00-01)

摘　　要：为了有效提升自由漂浮空间机器人的负载能力,提出一种基于多目标粒子群优化(multi-objective particle swarm optimization,MOPSO)算法的多约束多目标轨迹优化方法.结合建立的负载操作模式下空间机器人系统动力学模型,将负载最大化问题转化为实现关节力矩、基座扰动和系统能量同时最小的多目标轨迹优化问题;建立了相应的多目标优化问题(multi-objective optimization problem,MOP)数学模型;基于MOPSO算法求解出满足负载最大化要求的Pareto最优解集,并在算法中对约束条件进行了有效的处理.通过仿真实验证明了所提方法的有效性.In order to effectively increase the load carrying capacity of free-floating space robot, a multi-constrained multi-objective trajectory optimization method based on MOPSO （multi-objective particle swarm optimization） algorithm is proposed. Combined with the established dynamics model of space robot system under load carrying condition, load maximization is transformed into the multi-constrained multi-objective trajectory planning problem which simultaneously satisfy minimization of joint torque, base disturbances and system energy. The corresponding mathematical model of MOP （multi-objective optimization problem） is established. The Pareto solution set meeting the requirements of load maximization is solved by MOPSO algorithm which tackles constraint conditions effectively. By simulation, the availability of this method is proved.

关 键 词：空间机器人 负载最大化 多目标轨迹优化 多目标粒子群优化(MOPSO) 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]