Fuzzy Risk Evaluator for Collision Avoidance Design of Vessels Based on Automatic Identification System

Because of the high density of sea transportation, an effective solution for improving navigational safety is necessary. In this study, a warning system for the collision avoidance design of vessels is practically realized by integrating sensing messages delivered by an installed automatic identification system (AIS) and a fuzzy risk evaluator. Messages sent from the AIS include 1. the relative speed between any two selected vessels in the monitored open sea; 2. the distance between any two selected vessels; 3. the time of the closest point of approach, and 4. the distance of the closest point of approach. These four sensing messages extracted from the AIS are used as the inputs of the proposed fuzzy risk evaluator, and a fuzzy inference then makes an expert decision on the collision risk of any two selected vessels. The proposed system is programmed using the well-known software language C#, and a graphical user interface that can show the current positions of monitored vessels on Google Maps is also constructed. The proposed fuzzy risk evaluator can deliver real-time collision classification for all monitored ships, and the inference results can be used as reliable navigation guidelines for the collision avoidance design of vessels.