一类函数完全未知的随机非线性系统自适应H_∞跟踪控制  

Adaptive H-infinity tracking control for a class of stochastic nonlinear systems with completely unknown functions

作　　者：李小华[1] 刘辉 刘晓平 LI Xiao-hua;LIU Hui;LIU Xiao-ping(School of Electrons and Information Engineering,University of Science and Technology Liaoning,Anshan Liaoning 114051,China;Faculty of Engineering,Lakehead University,Thunder Bay ON,Canada)

机构地区：[1]辽宁科技大学电子与信息工程学院,辽宁鞍山114051 [2]湖首大学工程学院,加拿大安大略桑德贝

出　　处：《控制理论与应用》2019年第9期1431-1441,共11页Control Theory & Applications

基　　金：国家自然科学基金项目(61403177);辽宁省自然科学基金项目(20180550319);辽宁科技大学研究生教育改革与科技创新创业项目(LKDYC201701)资助~~

摘　　要：针对一类函数完全未知的严格反馈随机非线性系统,提出了一种基于backstepping技术的鲁棒H_∞自适应神经跟踪控制器设计的新方法.该方法可在随机非线性系统是依概率一致最终有界的情况下,保证随机非线性系统H_∞性能指标,且H_∞踪踪控制器容易获得.同时该方法去除了一些文献中神经网络逼近误差需要平方可积的假设.文中使用径向基函数(radial basis function, RBF)神经网络逼近打包的未知非线性函数.所设计的控制器能够保证闭环系统跟踪误差及其它所有信号都是依概率有界的,且对外界干扰具有鲁棒H∞抑制作用.最后,仿真结果验证了所提方法的有效性和正确性.In this paper,a novel approach based on backstepping technique is proposed to design an adaptive robust H-infinity tracking controller for a class of strict feedback stochastic nonlinear systems,in which all the functions are unknown.This method can guarantee that the controlled system has an H-infinity performance index under the condition that the stochastic nonlinear system is uniformly ultimately bounded in probability and the H-infinity tracking controller can be obtained easily.At the same time,based on the control scheme,the assumption that approximation errors must be square-integral in some literature has been eliminated.Radial basis function(RBF)neural networks are used to approximate the packaged unknown nonlinear functions.The designed controller can guarantee that the tracking error and other all the signals in the closed-loop system are bounded in probability,and the controlled system has an H-infinity disturbance attenuation performance for external perturbations.Finally,the simulation results are given to demonstrate the feasibility and validity of the proposed method.

关 键 词：随机系统 非线性系统 backstepping技术 H∞干扰抑制 鲁棒跟踪控制 自适应控制 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]