不确定非线性系统的模糊鲁棒跟踪控制  

Fuzzy Robust Tracking Control for Uncertain Nonlinear Systems

作　　者：刘亚[1] 胡寿松[1] 

机构地区：[1]南京航空航天大学自动化学院,南京210016

出　　处：《自动化学报》2004年第6期949-953,共5页Acta Automatica Sinica

基　　金：国家自然科学基金(60234010)高校博士点基金(20040287005)资助~~

摘　　要：提出了一种基于 T-S 模糊型的鲁棒自适应跟踪控制方法。整个控制方案在结合所有的局部线性状态反馈控制器的基础上，引入了基于自适应神经网络的鲁棒控制器，所提出的模糊自适应鲁棒控制器设计方法不需要求取李亚普诺夫方程的公共解，不要求系统的不确定性项满足任何匹配条件或约束条件．所提出的带有补偿项的完全自适应 RBF 神经网络，通过在线自适应调整 RBF 神经网络的权重、函数中心和宽度，提高丁神经网络的学习能力，可以有效地对消系统的未知不确定性的影响．同时通过自适应补偿项来在线估计神经网络的近似误差边界，弥补了神经网络的不足．所提出的方案保证了闭环系统的稳定性，有效地提高了系统的鲁棒性和跟踪性能．仿真实例表明了所提出方法的有效性．A robust adaptive tracking control method is presented based on T-S fuzzymodel.The overall control scheme is constructed by combining all local state feedbackcontrollers and robust adaptive neural network based controllers.By the proposed fuzzyadaptive robust controller there is no need to find a common positive definite matrix sat-isfying matrix Lyapunov equation,and no constraint or matching conditions are required.The learning ability of the proposed full adaptive RBF neural network with compensatingcomponent is improved through adaptive tuning of the weights,centers and widths on line.The adaptive compensating component is designed to compensate the shortcoming of theneural network through online estimating the bound on the neural network approximateerror.With the proposed method,the stability of the closed-loop system is guaranteed andimproved robustness and tracking performance are obtained.Simulation example is givento illustrate the effectiveness of the proposed method.

关 键 词：模糊T—S模型 RBF神经网络 自适应跟踪控制 鲁俸性 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置]