Structure-Aware Dynamic Graph Neural Networks for SSE
摘要
Autonomous vessels represent an important future direction for maritime transportation. Compared to traditional manned ships, autonomous vessels can reduce the probability of human error, enhance energy efficiency, and support more reliable navigation. However, their operation is significantly influenced by the marine environment. Sea waves induce motions such as pitch, roll, yaw, and heave, as well as coupled roll–pitch responses, which interact in complex ways and affect vessel safety and fuel consumption. Therefore, autonomous vessels must possess the capability to accurately identify surrounding sea states in real time.