Adaptive Fixed-Time Neural Tracking Control for Nonlinear Systems with Unknown Disturbances

Abstract

This paper investigates the adaptive fixed-time tracking control problem for nonlinear systems with unknown external disturbances. So as to estimate unknown disturbances, a fixed-time disturbance observer is introduced with the help of neural networks, where neural networks are utilized to approximate unknown nonlinear functions. By combining fixed-time theory with command filtered technique, an adaptive fixed-time control scheme is proposed, which not only evades the problem of "explosion of complexity" during the backstepping design process, but also ensures that the output tracking error could converge to a small neighborhood of the origin within fixed-time, and all signals in the closed-loop system remain bounded. Finally, the effectiveness of the designed control method is tested by a numerical simulation example.