An approach of vessel collision risk assessment based on the D–S evidence theory

Abstract

Computing time and decision correctness are the measurable indicators in vessel collision risk (CR) assessment. However, the existing CR assessment approaches, based on fuzzy theory or neural network, have lower accuracy and longer computing times. To overcome these drawbacks and obtain a compromised evaluation, an approach of vessel CR assessment based on the Dempster–Shafer (D–S) evidence theory is proposed in this paper. Considering that CR is associated with the membership functions corresponding to navigation parameters such as the distance to the closest point of approach (DCPA), the time to closest point of approach (TCPA) and the relative distance, we use the multiradar network to achieve them. Afterwards, applying the D–S evidence theory, we successfully assess CR with joint basic probability assignment (JBPA). Finally, the simulation results confirm the validity of the proposed approach.