基于PSO的神经网络机械臂自校正控制

作者：陈文元,杨东勇 日期：2008-02-25/span> 浏览：3230 查看PDF文档

基于PSO的神经网络机械臂自校正控制

陈文元,杨东勇
（浙江工业大学 信息工程学院，浙江 杭州 310014）

摘要：粒子群优化算法(PSO)基于群体的演化算法，本质上是一种随机搜索算法，并能以较大概率收敛到全局最优。针对非线性机械臂系统,利用径向基函数（RBF）神经网络和PID控制器作为混合控制器，运用PSO算法对神经网络参数进行在线学习优化，同时在PID控制器的辅助下对机械臂系统进行在线自校正控制。计算机仿真表明，该控制器具有较高的控制精度和响应速度，可以满足机械臂工作要求。
关键词：粒子群算法;径向基函数;神经网络;自校正控制
中图分类号：TP241文献标识码：A文章编号：1001-4551(2008)01-0044-04

Selftuning robotic system control based on PSO and neural network
CHEN Wenyuan, YANG Dongyong
（College of Information Engineering, Zhejiang University of Technology, Hangzhou 310014, China）
Abstract: Particle swam optimization (PSO) is an optimization technique based on evolutionary computation. The algorithm is a random searching algorithm in nature. It can converge to the global minima more probability. Aim to a nonlinear robotic arm system, a radial basis function (RBF) neural network and PID control strategy were proposed. PSO was used for optimizing parameters of neural network controller, selftuning control for the robot arm system was carried out by the PID controller. The simulation results show that the method is more precision and effective, it can content with the demand of robotic arm control.
Key words: particle swam optimization (PSO); radial basis function (RBF); neural network; selftuning control
参考文献(Reference):
［1］MAVROIDIS C, DUBOWSKY S, RAJU V. End Point Control of long Reach Manipulator Systems［C］//Proceedings of the Ninth World Congress on the Theory of Machines and Mechanisms, Portugal:［s. n］, September 1-3,1995, Milan, Italy,1995.
［2］YOSHIKAWA T, HOSODA K, TOSHITSUGU D, et al. Dynamic trajectory tracking control of flexible manipulator by macromicro manipulator system［J］. ICRA,1994:1804-1809.
［3］DAVID W C, DAVID P M, WAYNE J B. Experimental study on micromacro manipulator vibration control［J］. ICRA,1996:2549-2554.
［4］刘建昌,苗宇.基于神经网络补偿的机械臂轨迹控制策略的研究［J］.控制与决策,2005,20(7):732-735.
［5］宗光华.机器人技术手册［M］.北京:科学出版社,1996.
［6］ANGELES J.机器人机械系统原理理论、方法和算法［M］.宋伟刚,译.北京：机械工业出版社,2004.
［7］KENNEDY J, EBERHART R. Particle Swarm Optimization［C］//Proc IEEE Int Conf on Neural Network, Portugal: ［s. n.］, 1995:1942-194.
［8］曾建潮,崔志华.微粒群算法［M］.北京:科学出版社，2005.