Decision-Making For Multi-Ship Collision Avoidance Based On Clonal Selection Optimization

A new technology gradually developed by the analysis and improvement of decision-making for two ship collision avoidance is decision-making for multi-ship collision avoidance.A new idea is provided for multi-ship collision avoidance with the development of artificial intelligence algorithms.In this paper,the effectiveness of multi-ship collision avoidance is improved by applying the improved clonal selection algorithm to decision-making for multi-ship collision avoidance.In this paper,a method to calculate the composite collision risk index is proposed by using fuzzy mathematics on the basis of ship safe distance approach and the principle of ship collision geometry.The results of three different collision risk calculation methods are analyzed and compared.The results show that the method is more reasonable than the method of quantitative calculation of the general collision risk.The annealing strategy of simulated annealing algorithm is introduced into the clonal selection algorithm,and a clonal selection algorithm based on simulated annealing is designed.It combines the characteristics of clonal selection algorithm and simulated annealing algorithm in the optimization mechanism,structure and behavior,and makes up the respective weaknesses,so that the search ability of the two algorithms can be complemented.Examples are given to demonstrate the feasibility of the hybrid algorithm and compare the effectiveness of the proposed algorithm with the clonal selection algorithm.The relevant variables of the collision avoidance action constitute the string of the algorithm to establish reasonable multi-objective function by considering the factors influencing the collision avoidance in the study of multi-ship collision avoidance.And taking into account the differences between the various factors,the weight of objective function is linearly distributed to make the collision avoidance function model more in line with the actual navigation and simple comprehensive evaluation is carried by the affinity function,and the optimal solution to the problem is found by using iterative optimization of improved clonal selection algorithm.Finally,the optimization results are used to simulate the decision-making for multi-ship collision avoidance,and the validity of the avoidance action is verified.The results show that the algorithm can meet the requirements of real-time decision-making and provide collision avoidance support for officers.