基于粒子群优化算法的单腿机器人膝踝协调运动控制  

Coordinated Motion Control between the Knee and Ankle Joints for One-legged Robot Based on Particle Swarm Optimization Algorithm

作　　者：吴伟男[1,2] 朱秋国[1,2] 吴俊[1,2] 熊蓉[1,2] 

机构地区：[1]浙江大学智能系统与控制研究所,杭州310027 [2]浙江大学工业控制技术国家重点实验室,杭州310027

出　　处：《机械工程学报》2017年第15期93-100,共8页Journal of Mechanical Engineering

基　　金：国家自然科学基金(51405430;61473258;U150921);浙江省公益技术项目(2016C33G2010137)资助项目

摘　　要：针对三段式冗余单腿机器人的高能效运动问题,通过采集人体跳跃的运动学数据,分析人体在跳跃中膝踝关节的做功机理,提出冗余单腿机器人在跳跃中膝协调运动的定义,研究基于粒子群优化算法(Particle swarm optimization,PSO)搜索算法实现膝踝关节协调运动的优化方法,优化结果验证了膝踝关节协调运动规划算法的高能效性。通过仿真试验,实现了单腿机器人膝踝协调的连续跳跃运动,验证了规划算法和控制算法在跳跃中的有效性,通过实物试验验证了算法的有效性和鲁棒性。Aiming at the energy efficiency problem of three-segmented and redundant one-legged robot, the working principle of human＇s knee and ankle joints while hopping is analyzed by collecting and analyzing the kinematics data of human. The definition of coordinated motion between knee and ankle joints is proposed. Control algorithm based on particle swarm optimization（PSO） algorithm is also proposed to realize the coordinated motion between knee and ankle joints. Thus the energy-efficiency of kinematic coordinated planning algorithm is verified. The effectiveness of planning algorithm and control algorithm in the hopping is verified by simulation, implementing the coordinated continuous hopping movement between the knee and ankle joints. The effectiveness and robustness of the method is validated through the practical experiments.

关 键 词：单腿机器人 运动冗余 协调运动 粒子群优化算法 

分 类 号：TG156[金属学及工艺—热处理]