摘要
为了进一步提高机械臂在变负载下控制精度,设计了一种基于最优拉丁Widrow-Hoff网络的变负载机械臂自适应鲁棒控制方法。采用模态分解法创建关于机械臂的动力学状态方程,再以最优拉丁神经网络为基础提出具备自适应性能的鲁棒控制器,有利于加强机械臂在变负载情况下的系统鲁棒性以及控制精准度。仿真结果表明,利用最优拉丁Widrow-Hoff网络针对系统非线性实施估计,通过观测器系统总干扰实施补偿与估计,可以适应各种负载工况。试验验证文中所设计控制器具备较强的鲁棒性;可以全面考虑系统干扰以及系统非线性等各类因素,更契合控制的现实应用场景,具有更高的跟踪精准度。该研究有助于提高机器人的动作精度,为后续的特种环境的是适应性起到一定的推进作用。
In order to further improve the control accuracy of a variable load manipulator,an adaptive robust control method for a variable load manipulator based on the optimal Latin Widrow-Hoff network is designed.The modal decomposition method is used to create the dynamic state equation of the robot arm,and then a robust controller with adaptive performance is proposed based on the optimal Latin neural network,which strengthens the system robustness and control accuracy of the robot arm under varying loads.The simulation results show that the optimal Latin Widrow-Hoff network can be used to estimate the nonlinearity of the system and compensate and estimate the total interference of the observer system,which can be adapted to various load conditions.The experiment verifies that the controller designed in this paper has strong robustness.Various factors such as system interference and system nonlinearity are considered,which is more suitable for the realistic application scenarios of control,and tracking accuracy is improved.This research is helpful to improve the accuracy of the robot's action,and plays a certain role in promoting the adaptability of the subsequent special environment.
作者
黄贤振
阎兵
彭淑萍
HUANG Xianzhen;YAN Bing;PENG Shupin(School of Mechanical Engineering,Tianjin Polytechnic Normal University,Tianjin 300222,China;College of Manufacturing Engineering,Jiangxi Metallurgy Vocational and Technical College,Xinyu Jiangxi 338000,China)
出处
《机械设计与研究》
CSCD
北大核心
2023年第6期128-131,共4页
Machine Design And Research
基金
江西省教育厅科技项目(GJJ181423)。
作者简介
黄贤振(1989一),男,博士,讲师,主要从事智能制造技术,发表论文12篇,E-mail:p85236@126.com。
