Adaptive neural network tracking for a class of Markov jump stochastic nonlinear systems based on Extreme Learning Machine

Abstract

In this paper, the tracking problem is proposed for a class stochastic nonlinear systems with Markov jump based on neural network. By using backstepping method and Lyapunov function, the adaptive controller is designed to guarantee such the Markov jump stochastic nonlinear systems is asymptotically stable and track the desired reference model. The single hidden layer feed-forward neural network (SLFNN) is employed to approximate the unknown nonlinear functions, and the hidden layer node parameters are trained by extreme learning machine (ELM) algorithms. A simulation example demonstrates the effectiveness of the proposed method.