Research on adaptive neural networks spatial trajectory tracking control problem of an underactuated AUV under disturbances

Abstract

This paper studies the spatial trajectory tracking control issue for an underactuated autonomous underwater vehicle (AUV) with disturbances. Firstly, by performing coordinate transformation, a strict-feedback form of error variables is constructed, which facilitates the design of the scheme using the backstepping method. Secondly, neural networks are introduced to approximate the disturbances and the complex dynamic of AUV, which not only reduces the requirement for model accuracy in the backstepping method, but also simplifies the design process. At the same time, the inevitable problem of “computational explosion” in the backstepping method is solved by the command filter, and combined with the adaptive control method, the weights of the neural networks are adjusted online to complete the design of the adaptive neural networks command filter controllers. Lyapunov theory analysis demonstrates that all signals are ultimately bounded, and the tracking error converges to a small neighborhood near the origin. Finally, a comparative simulation is provided to validate the effectiveness and superiority of the designed control scheme.