Fuzzy-Logic based Ship-Bridge Collision Alert Model Form Ship Behaviour Perspective

Ship collision with offshore structures such as bridges poses potential serious risk. This paper proposes a fuzzy-logic based ship-bridge collision risk alert model from the perspective of ship behavior using Automatic Identification System data. The kernel of the proposed method is to identify the risk factors from the historical data by comprehensively considering the traffic-flow direction, water level seasons, course of ground and speed of ground, to fuzzify the risk factors by using ship behaviors, fuzzy rule base to establish the fuzzy logic boxes and finally to obtain the collision risk. The result demonstrates that the proposed model cannot only predict the collision risk with good accuracy but also be used to improve the ship handling when passing under the bridge. Consequently, this paper proposes a practical and beneficial model for ship-bridge collision alert system.