A Navigation Method for UUVs under Ocean Current Disturbance Based on Deep Reinforcement Learning

Unmanned Underwater Vehicles (UUVs) are wired or wireless unmanned vehicles that sail underwater by remote control or automatic control. Compared with submarines, UUVs have demonstrated strong stability, risk tolerance and flexibility, and can dive into deep sea to perform various tasks. In order to further improve the adaptability of UUVs under ocean current disturbance, a navigation method based on deep reinforcement learning is proposed in this paper, which considers both static and dynamic environmental factors. The effectiveness of the proposed UUV navigation method was evaluated and analyzed through simulation experiments.