Course Tracking Control of USV Based on Optimal Backstepping Method Via Reinforcement Learning

In this paper, an optimal control problem based on optimal backstepping (OB) method is studied for the ship course tracking system with unknown dynamics. Firstly, reinforcement learning (RL) technology is employed to deal with the optimal control problem. Secondly, due to the unknown dynamics of ship course tracking system, the neural network is applied in the process of design. Thirdly, in order to ensure the result of the whole system is optimal, the OB method is developed. Finally, the stability of the closed loop system is proved based on Lyapunov stability theory and the simulation further illustrated the effectiveness of the proposed algorithm.