Data-Driven Ship Stay Behavior Identification in Maritime Internet of Things System

With the rapid developments of connected devices, communication networks, and the need for improved efficiency and safety in the maritime domain, A large amount of ship data will be generated in the Maritime Internet of Things (IoT) system. Extracting ship stay behavior from these data has become an important issue, as it can provide a guarantee for safe vessel navigation and enhance the efficiency of maritime supervision. This paper focuses on ship trajectory data and constructs a semantic trajectory model for ships to enrich the data and achieve a semantic representation of ship behavior. Based on ship trajectory data, a method for identifying trajectory stay points is proposed, which combines ship trajectory local motion characteristics to achieve ship stay behavior recognition. Parameter tuning experiments are conducted, and a comparative experiment is conducted using vessel data from a 6-month period in the South China Sea. The results show that the proposed method achieves good performance in terms of robustness and accuracy.