Ship collision situation assessment based on GA-BP and HMM

To proactively identify collision risks in ship navigation and enhance maritime traffic safety, a novel method for evaluating ship collision risk status using GA-BP and HMM algorithms is introduced. Initially, an analytic hierarchy process is employed to develop a ship collision risk assessment model. Subsequently, historical trajectory data of ships along with pertinent environmental information are leveraged to train the BP and its optimization algorithm, enhancing the prediction accuracy for potential ship collisions. Furthermore, the HMM model is utilized to evaluate ship encounter situations. The results demonstrate that this approach significantly improves the accuracy and reliability of ship collision risk assessments, providing critical decision-making support to shipping companies and crew members in scenarios of collision risk.