Attitude tracking for high-speed aircraft based on discrete adaptive neural network backstepping control under unknown disturbances

Unknown disturbances can significantly degrade the attitude tracking performance of high-speed aircraft and may even threaten system stability. To mitigate the effects of external disturbances, this article proposes a discrete neural network adaptive backstepping controller based on gradient descent algorithm, aiming to improve the attitude tracking accuracy of high-speed aircraft. First, the model of high-speed aircraft is transformed into an equivalent predictive model in a strict-feedback form to address the coupling problem of the system. Subsequently, discrete neural network adaptive backstepping controller is designed based on the transformed predictive model. To effectively handle uncertainties in the equivalent predictive model, a radial basis function neural network is utilized to estimate the unknown nonlinearities. A distinguishing feature of the proposed method is its parameter update mechanism based on gradient descent, which simultaneously adjusts the weights, centers, and widths of the neural network. The stability of system is rigorously proven by Lyapunov stability theory. Finally, simulation experiments are conducted to verify the effectiveness and robustness of discrete neural network adaptive backstepping controller.