严格反馈非线性系统的自适应神经网络输出反馈控制  

Adaptive neural output feedback control for strict feedback nonlinear system

作　　者：孙国法[1] 田宇[2] 王素珍[1] SUN Guo-fa;TIAN Yu;WANG Su-zhen(College of Automation Engineering, Qingdao University of Technology, Qingdao Shandong 266520, China;Beijing Aerospace Automatic Control Institute ,National Key Laboratory of Science and Technology on Aerospace Intelligence,China Academy of Launch Vehicle Technology, Beijing 100854, China)

机构地区：[1]青岛理工大学自动化工程学院,山东青岛266520 [2]北京航天自动控制研究所,宇航智能控制技术国家级重点实验室,北京100854

出　　处：《控制理论与应用》2017年第3期375-382,共8页Control Theory & Applications

基　　金：中国博士后科学基金项目(2013M541939);国家自然科学基金项目(61640302)资助~~

摘　　要：基于非线性反馈函数,文章设计神经网络状态观测器,解决一类非线性系统的输出反馈控制问题.非线性反馈神经网络观测器在系统存在不确定性函数的情况下实时估计系统状态.利用所获得的状态信号,设计了自适应神经网络动态面控制器,同时保证了闭环系统的稳定性和所有信号的有界性.通过调节设计参数的取值能够达到期望的闭环跟踪性能.数值仿真表明,所设计的状态观测器不需要对原系统做状态变换,能够克服输出反馈滑模控制器带来的抖震问题.Based on nonlinear feedback function,this paper investigates output feedback control problem for a class of nonlinear systems.In the case of systems with uncertain functions,nonlinear feedback neural network observer estimates system states in real time.Employing the obtained state signals,an adaptive neural network controller is designed to guarantee the stability of the closed-loop system and the boundedness of all signals.The desired closed-loop tracking performance can be achieved by adjusting the value of the design parameters.Numerical simulation results show that,without state transformation,the state observer overcomes the chattering problem caused by the output feedback sliding mode controller.

关 键 词：状态观测器 动态面控制 自抗扰控制 神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP273[自动化与计算机技术—控制科学与工程]