Neuroadaptive Impulsive Control of a Class of Uncertain Nonlinear Systems

Abstract

This paper addresses the tracking control problem for a class of uncertain nonlinear systems, offering a neuroadaptive impulsive control solution with fast weight convergence. The proposed control architecture uniquely employs radial basis function neural networks (RBFNNs) to approximate unknown system dynamics, where the neural network (NN) weight estimator is updated at each impulsive instant. Unlike existing neuroadaptive controllers, this method quickly identifies unknown dynamics without inducing high-frequency oscillations due to the impulse update of the NN weight estimator, thereby improving the transient performance of the closed-loop system. Leveraging the Lyapunov stability theory for impulsive dynamical systems, this paper rigorously establishes the semi-global uniform ultimate boundedness (SGUUB) of all closed-loop system signals. Finally, validation through simulation studies substantiates the efficacy of the proposed neuroadaptive impulsive controller.