COLLISION AVOIDANCE PATH PLANNING FOR SHIPS BY PARTICLE SWARM OPTIMIZATION

Ship collision avoidance is a key consideration in maritime systems. Collision avoidance maneuvers depend on navigators' experience and skill levels. Because both maritime traffic densities and average ship speeds are increasing, the times available for decision-making are decreasing, which elevates the risk of human errors in the collision avoidance process. To reduce the effect of human factors and efficiently prevent collisions between ships navigating in open water with effective visibility, a particle swarm optimization (PSO) algorithm can be used to plan ship paths. An improved ship domain dynamic model can assess collision risks in close-range encounters. Several marine traffic scenarios based on standard encounter types were simulated; the proposed PSO algorithm was tested in those scenarios. This paper discusses the compatibility and consistency of the algorithm outputs as well as the execution efficiency of the algorithm.