基于神经网络补偿的机械臂轨迹控制策略的研究_刘建昌.pdf

第20 卷第7 期 　控　制　与　决　策 2005 年7 月 Vol. 20 No .7 　Control and Decision　 July 2005 : 1001-0920(2005) 07-0732-05 刘建昌, 苗　宇 ( , 110004) : 建立了二维机械臂的动力学方程, 分析了控制机械臂轨迹应考虑的不确定 . 考虑到机械臂动力学模型的 非线性和参数的不确定性, 提出了采用神经网络作为补偿器的机械臂轨迹控制策略. 该策略结构简单, 采用标称值确 定神经网络初始权值, 易于工业实现. 仿真研究表明了所提控制策略的有效性及神经网络补偿器的泛化能力和自适 应能力. : 二维机械臂; 神经网络; 补偿器; 轨迹控制 : T P24: A Research on Traj ectory Control Strategy of Robot Arm Based on Neural Network Compensation LI U Jian-chang, MIA O Yu ( Key La oratory of Process Industry Automation, Minist ry of Education, Nort heast ern University , Shenyang 110004, China. Correspondent: LIU Jian-chang, E-m ail: jcliu@ mail. neu. edu. cn) Abstract: T he dynamics equations of the tw o dimensional ro ot arm are esta lished, and the uncert ain varia les . which should e considered t o cont rol t he trajectory of ro ot arm are analyzed The compensating control strategy ased on neural net works is presented,w hich considers the the nonlinearity of dynamics models and the uncertainty of parameters. T he structure of the presented control st rategy is simple, and the nominal value is used to fix the - , . initial w eight of neural net work which is easy to realize in industry T he simulation research show s the , . effect iveness of the control strategy and the general and adaptive a ility of neural netw ork compensators Key words: T w o dimensional ro ot arm; Neural netw ork; Compensator; Trajectory control 1 引　　言 ) , , , ,