摘要
针对机械臂轨迹跟踪控制中传统滑模控制需估计其建模误差及外界干扰等不确定性,当建模不确定性及外界干扰较大较复杂时,将会导致出现抖振现象。该文在以传统滑模控制为主控制器的基础上,通过对传统干扰观测器进行改进,对外界干扰进行反馈补偿,同时利用神经网络对其建模误差进行逼近。通过机械手仿真实验结果表明,所提方法能够有效抑制系统抖振现象,提高响应速度及其轨迹跟踪精度。
For the traditional sliding mode control in the trajectory tracking control of the robotic arm,it is necessary to estimate its modeling error and external interference and other uncertainties.When the modeling uncertainty and external interference are large and complex,it will cause chattering.On the basis of traditional sliding mode control as the main controller,the algorithm improves on the basis of traditional interference observers,feedback compensates for external interference,and uses neural networks to approximate its modeling errors.The simulation results of the manipulator show that the proposed method can effectively suppress the chattering of the system and improve the response speed and the tracking accuracy of the trajectory.
作者
许崇立
王冬青
周振
XU Chongli;WANG Dongqing;ZHOU Zhen(School of Automation,Qingdao University,Shandong Qingdao 266071,China;School of Electrical Engineering,Qingdao University,Shandong Qingdao 266071,China)
出处
《工业仪表与自动化装置》
2021年第1期3-7,12,共6页
Industrial Instrumentation & Automation
基金
国家自然科学基金资助项目“复杂网络拓扑与参数辨识”(61873138)
国家自然科学基金资助项目“基于数据特征的多模态过程辨识建模方法”(61573205)。
关键词
不确定性
滑模控制
改进干扰观测器
神经网络
uncertainty
sliding mode control
improved disturbance observer
neural network
