Regulation for Nonlinear Systems Using Recurrent High-Order Neural Network Under Aperiodic Measurements

Abstract This paper presents a solution to the output regulation problem for nonlinear systems with unknown dynamics, when the states are aperiodically measured. Our methodology overcomes conventional limitations by utilizing a recurrent high-order neural network (RHONN) as an identifier. It employs a new impulsive adaptive algorithm to train the weights online at measurement instants. This methodology allows trajectory tracking with bounded error by solving the Francis–Byrnes–Isidori equations for the RHONN model and using the solution to develop a controller for the nonlinear system identified. The stability of identification and tracking errors are analyzed. Simulations have been included to demonstrate the robustness and effectiveness of the proposed technique.