随机多输入多输出系统的自适应神经网络控制

针对一类具有严格反馈形式的随机非线性多输入多输出系统的自适应神经跟踪控制问题,本文利用径向基函数神经网络的万能逼近性,结合自适应Backstepping设计方法,提出了一类新的自适应神经网络状态反馈控制器,并对该系统提出的控制器含有较少的参数问题,通过Lyapunov稳定性理论进行了稳定性分析和证明,并应用仿真算例进行验证,仿真结果表明,闭环系统的所有误差变量概率意义下有界,并使系统的输出收敛到参考信号的一个小的邻域范围之内。该研究对随机非线性多输入多输出系统的跟踪控制有一定的指导意义。

Adaptive Neural Tracking Control for Stochastic Nonlinear MIMO Systems

This paper addresses the problem of adaptive neural control for a class of stochastic nonlinear multi-input and multi-output （MIMO） systems with strict-feedback form.The new controller of adaptive neural network with state feedback is presented by using a universal approximation of radial basis function neural network and backstepping.The problem that controller of the system contains less parameters has been analyzed and the stability has been proved by the Lyapunov stability theory.Numerical example is given for illustration.The simulation results show that all the variables in the closed-loop system are stochastic bounded while the system output tracking the desired reference signal and the tracking error converges to a small enough neighborhood of origin.The research for tracking conrtol of stochastic nonlinear multiinput and multi-output systems has certain guiding significance.