Fuzzy Neural Network-Based Adaptive Control for a Class of Uncertain Nonlinear Stochastic Systems

TLDR

The paper addresses nonaffine pure‑feedback nonlinear stochastic systems, a class not previously controlled under stochastic disturbances. The study develops an adaptive tracking controller for nonlinear stochastic systems with unknown dynamics. Fuzzy‑neural networks approximate the unknown functions, and a backstepping design yields the controller and adaptation laws, which are validated by simulation. The controller requires fewer tunable parameters, reducing computational load, and Lyapunov analysis guarantees semiglobal uniform ultimate boundedness in probability with bounded tracking error.

Abstract

This paper studies an adaptive tracking control for a class of nonlinear stochastic systems with unknown functions. The considered systems are in the nonaffine pure-feedback form, and it is the first to control this class of systems with stochastic disturbances. The fuzzy-neural networks are used to approximate unknown functions. Based on the backstepping design technique, the controllers and the adaptation laws are obtained. Compared to most of the existing stochastic systems, the proposed control algorithm has fewer adjustable parameters and thus, it can reduce online computation load. By using Lyapunov analysis, it is proven that all the signals of the closed-loop system are semiglobally uniformly ultimately bounded in probability and the system output tracks the reference signal to a bounded compact set. The simulation example is given to illustrate the effectiveness of the proposed control algorithm.