基于PID迭代学习控制的码垛机器人研究  

Research on Palletizing Robot Based on PID Iterative Learning Control

作　　者：于镭[1] 徐理达 YU Lei;XU Lida(School of Automation and Electronic Engineering,Qingdao University of Science andTechnology,Qingdao Shandong 266042,Chin)

机构地区：[1]青岛科技大学自动化与电子工程学院,山东青岛266042

出　　处：《机床与液压》2018年第11期70-74,共5页Machine Tool & Hydraulics

摘　　要：以平行四杆件的码垛机器人为研究对象,运用机器人动力学中的拉格朗日功能平衡法建立码垛机器人的动力学模型。在经典PID控制的基础上,运用迭代学习控制(ILC)对系统未知参数和干扰进行补偿,实现机器人系统各关节对期望运行轨迹的跟踪控制。通过对控制系统的分析和仿真可得,在机器人系统稳定工作的前提下,将迭代学习率设置为0.75,经过10次迭代学习之后,系统的误差绝对值最大值与期望轨迹的比值能够控制在10%以下。实验结果表明,码垛机器人的PID迭代学习轨迹跟踪控制可以获得较高的跟踪精度。A four-bar parallel palletizing robot is used as the object of research. The dynamic model of palletizing robot is established by using the Lagrange functional equilibrium method in robot dynamics. On the basis of classical Proportion Integral Derivative（ PID） control,Iterative Learning Control（ ILC） was used to compensate the unknown parameters and disturbances of the system,and the tracking control of the desired trajectory of each joint of the robot system was realized. Through the analysis and simulation of the control system,under the premise of stable operation of the robot system,the iterative learning rate was set to 0. 75. After 10 iterations,the ratio of the maximum absolute value of the system error over the desired trajectory could be controlled below 10%. The experimental results show that the PID iterative learning tracking control of the palletizing robot can obtain quite high tracking precision.

关 键 词：平行四杆件码垛机器人 拉格朗日功能平衡法 PID迭代学习控制 轨迹跟踪控制 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]