A Study on Collision Avoidance Timing for Autonomous Ships using Machine Learning

초록

To ensure the safe navigation of Maritime Autonomous Surface Ships (MASS), it is crucial to establish a robust collision avoidance system. In this study, we designed a system for autonomous vessels to avoid collisions with other ships. Utilizing the Decision tree model, a machine learning algorithm, we wanted to derive a rule to classify the collision avoidance points determined in AIS data. By applying the optimal collision avoidance timing obtained through this process to the collision avoidance system and validating it through simulation, we confirmed the system's performance in determining collision avoidance timing equivalent to that of human decision-making. © 2024 by the International Society of Offshore and Polar Engineers (ISOPE).

키워드