3D trajectory tracking control of an underactuated AUV based on adaptive neural network dynamic surface

Abstract

This paper addresses the 3D trajectory tracking control of an underactuated autonomous underwater vehicle (AUV) under uncertain model parameters and unknown external disturbances. A dynamic surface control scheme based on neural network and adaptive technique is proposed. In controller design, the first-order integral filters are employed to estimate derivative of virtual control, which avoid repeated derivative of virtual control. To deal with the effect of unknown external disturbances and uncertain model parameters, the neural network and adaptive technique are combined to approximate unknown nonlinear functions. All of the error signals in the closeloop system are uniformly ultimately bounded based on Lyapunov stability theory. Simulation studies and comparisons with adaptive dynamic surface control scheme illustrate the effectiveness and superiority of the proposed control scheme.