关键词: 自主水下航行器（AUV）, 姿态控制, 反步法, 神经网络, 滑模控制, 鲁棒性

Abstract: For the attitude control problem of Autonomous Underwater Vehicle（AUV） in automatic cruise task, a robust adaptive attitude control algorithm based on neural network and sliding mode control is proposed. The RBF neural network is used to approximate the uncertain terms in the mathematical model of AUV, and the influence of unmodeled dynamics and parameter perturbation is suppressed. Then, the attitude control law is designed based on the backstepping and sliding mode control, and the robust term is introduced to overcome the external disturbance and approximation error of neural network. Meanwhile, the stability of the control system is proved by Lyapunov theorem. The proposed control algorithm is applied to the attitude control system of an AUV for numerical simulation, and the effectiveness and robustness of the control algorithm are verified.

Key words: Autonomous Underwater Vehicle（AUV）, attitude control, backstepping, neural network, sliding mode control, robustness