DEVELOPMENT OF COLLISION AVOIDANCE ALGORITHM BASED ON CONSCIOUSNESS OF SHIP OPERATOR

Several collision avoidance algorithms employing various methods have been developed to date. In this paper, a collision avoidance algorithm is proposed based on the potential assessment of risk (PARK) model, which is a maritime traffic risk assessment model based on the characteristics of the Korean coastal regions and the consciousness of a ship's operator. The proposed collision avoidance algorithm is intended to be used in the transition period during which autonomous and human-operated ships sail simultaneously. The algorithm avoids collisions using the PARK model that reflects the International Regulations for Preventing Collisions at Sea. Situations such as head-on, crossing, overtaking, and their combination were simulated. The performance of the collision avoidance algorithm was verified by successfully avoiding collisions while simulating a busy waterway using historical automatic identification system data.