基于高阶内模迭代学习的关节机器人控制研究

Research on Articulated Robot Control Based on High-order Internal Model Iterative Learning

为了提升关节机器人在非严格重复条件下工作过程中的跟踪精度及响应速度,设计了三关节机器人模型,进行运动学及动力学分析来验证模型结构合理。针对关节机器人系统非重复性、非线性的特点,提出将高阶内模迭代学习控制算法应用于关节机器人系统控制中,设计合理的学习增益及更高的内模阶数,从理论上严格证明其收敛性。设计了仿真对比实验及加入扰动后的轨迹跟踪实验,结果表明,高阶内模迭代学习算法收敛速度更快并且具有良好的控制效果。

In order to improve the tracking accuracy and response speed of the articulated robot in the working process under non-strict repetitive conditions,a three-joint articulated robot model is designed,and the kinematics and dynamics analysis are carried out to verify the reasonable structure of the model.In view of the non-repetitive and nonlinear characteristics of the articulated robot system,it is proposed that a high-order in⁃ternal model iterative learning control algorithm can be applied to the control of the articulated robot system.A reasonable learning gain and a higher internal model order are designed to strictly prove its convergence in theo⁃ry.The simulation contrast experiment and the trajectory tracking experiment after adding the disturbance are designed.The results show that the high-order internal model iterative learning algorithm converges faster and has good control effect.