Dynamic Surface Control Using Neural Networks for a Class of Uncertain Nonlinear Systems With Input Saturation

TLDR

The paper proposes a dynamic surface control scheme for uncertain strict‑feedback nonlinear systems with input saturation and unknown disturbances, incorporating a nonlinear disturbance observer. The method employs a radial basis function neural network to approximate the unknown dynamics and a nonlinear disturbance observer, integrating them into a dynamic surface control design based on backstepping. Simulations confirm that the proposed DSC with RBFNN and NDO achieves bounded convergence of all signals, relaxes disturbance boundary assumptions, and avoids backstepping complexity.

Abstract

In this paper, a dynamic surface control (DSC) scheme is proposed for a class of uncertain strict-feedback nonlinear systems in the presence of input saturation and unknown external disturbance. The radial basis function neural network (RBFNN) is employed to approximate the unknown system function. To efficiently tackle the unknown external disturbance, a nonlinear disturbance observer (NDO) is developed. The developed NDO can relax the known boundary requirement of the unknown disturbance and can guarantee the disturbance estimation error converge to a bounded compact set. Using NDO and RBFNN, the DSC scheme is developed for uncertain nonlinear systems based on a backstepping method. Using a DSC technique, the problem of explosion of complexity inherent in the conventional backstepping method is avoided, which is specially important for designs using neural network approximations. Under the proposed DSC scheme, the ultimately bounded convergence of all closed-loop signals is guaranteed via Lyapunov analysis. Simulation results are given to show the effectiveness of the proposed DSC design using NDO and RBFNN.