Adaptive DDPG Design-Based Sliding-Mode Control for Autonomous Underwater Vehicles at Different Speeds

Autonomous underwater vehicles (AUVs) are becoming increasingly popular for ocean exploration, military and industrial applications. Motion control of AUV is the key to completing these missions. Sliding mode control (SMC) method has a good performance in the motion control system of AUV. Nevertheless, when the system is not fully or previously known, classical techniques are not entirely suitable for the SMC tuning. Moreover, the variation of drag and lift coefficients of the AUV are very sensitive to the AUV speed, and it is difficult to achieve the accurate control of AUV motion by fixed SMC parameters at different speeds. In this paper the tuning process at different speeds is automated through the use of a model-free reinforcement learning algorithm-deep deterministic policy gradient (DDPG) based SMC. The robustness and effectiveness of the proposed control method are tested and validated through extensive simulation results. The results show that the SMC-DDPG achieve motion control at different AUV speeds with fine stability, fast convergence speed, high precision and little chattering.