ABSTRACT This study proposes a neural network‐based adaptive dynamic surface control (DSC) approach for high‐order strict‐feedback nonlinear systems influenced by unmodeled dynamics, time‐varying input delays, and unknown backlash‐like hysteresis. Radial basis function neural networks (RBFNNs) are employed to approximate unknown system functions, facilitating effective compensation for uncertainties. A dynamic signal is introduced to counteract the effects of unmodeled dynamics, while the known upper bound of the input delay is utilized for delay compensation. To ensure stability in high‐order systems, a Lyapunov‐based function is formulated, and the integral mean value theorem is applied to handle time‐varying delays. Furthermore, an inequality transformation method, along with virtual controllers, is developed to reduce control errors induced by external disturbances. The proposed control scheme guarantees that the tracking error remains within a small neighborhood of the origin, with all system signals maintaining semi‐globally uniformly ultimately bounded (SGUUB) behavior. Simulation results confirm the effectiveness and robustness of the proposed approach.
Kharrat et al. (Fri,) studied this question.