基于OBL-SAMPSO算法的机器人零力控制摩擦力项辨识方法  

Identification Method of Friction Force Parameter in Robot Force-free Control Based on OBL-SAMPSO Algorithm

作　　者：王思源 张秋菊 WANG Siyuan;ZHANG Qiuju(School of Mechanical Engineering,Jiangnan University,Wuxi 214122)

机构地区：[1]江南大学机械工程学院,无锡214122

出　　处：《计算机与数字工程》2024年第7期1974-1980,共7页Computer & Digital Engineering

摘　　要：针对机器人零力控制中摩擦模型参数辨识,提出一种基于反向学习(OBL)策略的改进模拟退火(SA)寻优算法(OBL-SAMPSO算法)进行LuGre摩擦模型参数辨识。仿真结果表明:相较于传统的PSO算法,OBL-SAMPSO算法使参数辨识的绝对误差平均下降了85.86%;基于单关节实验平台进行LuGre摩擦辨识,并应用到其关节零力控制策略中,实现了具有良好柔顺性的拖曳操作。An improved simulated annealing optimization algorithm(OBL-SAMPSO algorithm)based on reverse learning strategy is proposed for parameter identification of LuGre friction model in robot force-free control.The simulation results show that compared with the traditional PSO algorithm,the OBL-SAMPSO algorithm reduces the absolute error of parameter identification by 85.86%on average.LuGre friction identification is carried out based on the single joint experimental platform,and applied to the joint force-free control strategy.The towing operation with good compliance is realized.

关 键 词：OBL-SAMPSO算法 零力控制 LuGre摩擦 参数辨识 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]