Ship collision avoidance is the core of intelligent ship research,and the objective evaluation of the avoidance performance of collision avoidance algorithms is its key link.Aiming at the problem of inaccurate and incomplete evalua-tion of ship collision avoidance algorithms,a method for evaluating the avoidance performance of ship collision avoidance algorithms based on the theory of topological cloud is constructed.Based on the International Regulations for Collision A-voidance at Sea,a ship collision avoidance algorithm avoidance performance evaluation index system is constructed,and the combined method of hierarchical analysis and entropy weighting is used to determine the comprehensive weights of the in-dicators for ship collision avoidance algorithm performance evaluation,and further establish the standard cloud object ele-ment model for ship collision avoidance algorithm performance evaluation,and determine the algorithm performance level by the method of cloud correlation calculation.Finally,a large number of encounter scenarios are randomly generated based on Monte Carlo ideas to evaluate the avoidance performance of different algorithms,and the evaluation results are verified.The results show that the proposed method can evaluate the avoidance performance of collision avoidance algo-rithms more accurately and reliably,and can reflect the advantages and disadvantages of different algorithms applied to different encounter situations.
ship collision avoidanceevaluation index systemtopological cloudAHP methodentropy weight method
NETL
NSTL
万方数据
