As the core equipment for in situ deep-sea scientific research, the design quality and operational performance of the sampling platform determine the efficiency of deep-sea operations, including long-term continuous observation, high-quality sampling and preservation, and in situ experimental studies. To address technical bottlenecks commonly observed in conventional platforms, including track sinkage, excessive motion drag, a...

This paper proposes the Risk-Aware SAC (RA-SAC) algorithm for unmanned surface vehicle (USV) path planning in the presence of high-speed dynamic obstacle interference. This method integrates risk perception, obstacle motion state extrapolation and heading constraint reward into the SAC framework. While preserving the entropy-maximization mechanism of SAC, it enhances long-horizon obstacle avoidance awareness through reward sha...