A Probabilistic Decision-Making System for Joining Traffic Lanes Within an Inland Traffic Separation Scheme

Abstract

Abstract The timing of joining traffic lanes is significant for safety of maritime transportation within a traffic separation scheme (TSS), which is emphasized in the International Regulations for Preventing Collisions at Sea. This paper proposes a probabilistic decision-making approach for a ship joining traffic lanes within a TSS that considers ship intention, ship characteristics, and traffic situation. It can be implemented in a decision support system for safe navigation or be included in the process of autonomous navigation. The kernel of this proposed approach is first to analyze the crossing and converging behavior in scenarios of joining near-lane and opposite-lane traffic, to determine the graphical structure after identifying the influencing factors (e.g., ship speed, position, and traffic condition), and then to quantify the model based on the geometrical relationships, logical relationships, and arithmetic relationships among influencing factors. The proposed approach is applied to two real scenarios of a ship joining traffic lanes, and the result consistently agrees with the ship strategy of joining traffic lanes from the historical trajectory. Consequently, this paper proposes a practical and useful tool to decide the timing of joining traffic lanes in busy waterways.