Adaptive prescribed performance second-order sliding mode tracking control of autonomous underwater vehicle using neural network-based disturbance observer

For trajectory tracking control problem of autonomous underwater vehicle (AUV) with model uncertainties, unknown nonlinear external disturbances, and input saturation, an observer-based adaptive second-order sliding mode control scheme with prescribed performance is proposed. Firstly, a finite time performance function (FTPF) is constructed to constrain the tracking error to the prescribed precision within the preset finite time, and achieve the prescribed dynamic convergence performance and tracking accuracy. Then, a neural network-based disturbance observer (NNDO) is designed to deal with model uncertainties and external disturbances, respectively. Based on prescribed performance control and sliding mode control technique, an adaptive prescribed performance second-order sliding mode control strategy was proposed. In addition, we construct an auxiliary system to overcome the effect of thrust saturation. Lyapunov method is applied to demonstrate the stability of the closed-loop system. Finally, the validity of the proposed control law is verified by numerical simulations. • A finite time performance function (FTPF) is introduced for AUV finite time control. • The arctan-type function is introduced to combine the prescribed performance method with the sliding mode technique. • An improved neural network-based disturbance observer is designed to accurately compensate unknown factors.