改进狼群算法的SCARA机械臂自适应迭代学习控制  

Adaptive Iterative Learning Control of SCARA Manipulator with Improved Wolf Pack Algorithm

作　　者：张长胜[1] 马泽楠 李宽[1] 陈标发 李伟[1] 杨俊[1] Zhang Changsheng;Ma Ze′nan;Li Kuan;Chen Biaofa;Li Wei;Yang Jun(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650550,China;Guangzhou Kossi Intelligent Technology Co.,Ltd.,Guangzhou 510000,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,昆明650550 [2]广州科伺智能科技有限公司,广州510000

出　　处：《机械科学与技术》2020年第8期1177-1183,共7页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金项目(51665025)资助。

摘　　要：针对减小机器人重复运动的位置、速度跟踪误差的问题,给出一种基于狼群算法优化的机械臂自适应迭代学习控制策略。根据SCARA(Selective compliance assembly robot arm)机械臂驱动方程,设计动力学系统的迭代学习控制律。引入自适应步长的狼群算法,使狼群能够根据猎物气味浓度动态调整移动步长,提高了算法的收敛速度和精度。该策略对机械臂控制器参数KP、KD进行寻优时,得到了良好的控制效果,实现了对期望轨迹的有效跟踪。实验结果表明,该算法灵活性好,对系统期望轨迹具有较高的跟踪精度,有效降低了双关节机械臂的位置、速度跟踪误差,具有较强的可行性与有效性。Aiming at reducing the position and speed tracking error of robot repetitive motion,an adaptive iterative learning control strategy based on wolf group algorithm is presented in this paper.According to the SCARA(Selective compliance assembly robot arm)manipulator drive equation,the iterative learning control law of the dynamic system is designed.The wolf group algorithm with adaptive step size is introduced to enable the wolf to dynamically adjust the moving step size according to the prey odor concentration,which improves the convergence speed and accuracy.When the strategy optimizes the parameters of the arm controller KP and KD,the result is effect for control and achieves effective tracking of the desired trajectory.The experimental results show that the adaptive control systems has good flexibility,high tracking accuracy for the system's desired trajectory,and effectively reduces the position and velocity tracking error of the double joint manipulator,which has strong feasibility and effectiveness.

关 键 词：SCARA机械臂 自适应迭代学习控制 狼群算法 自适应步长 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]