A Collision Risk Identification Method for Autonomous Ships Based on Field Theory

Collision risk identification is necessary for autonomous ships to recognize the risk when facing different encounter situations and make the proper collision avoidance decision. For achieving quantitative assessment of collision risk, this paper proposed a collision risk identification method based on field theory. Specifically, the method first combines the uncertainty model of ship position prediction to improve the quaternion ship domain (QSD). Then, based on the field theory and the situation of encounters, the ship risk field is constructed, which transform the ship domain overlapping index into the field energy superposition problem. The validation and superiority of this approach is examined by simulation studies. The results demonstrate that the proposed method can overcome some deficiencies of the conventional risk assessment methods. Moreover, the approach can effectively identify the collision risk in real time and provide a reference on furthering enhance the navigational safety for the autonomous ships. Therefore, the approach gives it a significant potential for use in collision risk identification in the future.