Research On Ship Collision Avoidance Based On Fuzzy Theory And Genetic Algorithm

As China’s economy rapidly develops and foreign trade prospers,maritime transportation has gradually become one of the irreplaceable modes of transportation.However,with the continuous increase in the number and density of ships,the frequency of ship collisions has also risen.To reduce the incidence of ship collisions,many experts and scholars have devoted themselves to the study of ship collision avoidance.However,most current research on ship collision avoidance focuses on the relative motion between ships,while ignoring the impact of external environmental factors and the inherent properties of ships.Therefore,this thesis combines these two points to conduct research on ship collision avoidance based on fuzzy theory and genetic algorithm.Firstly,a method for constructing a composite ship domain using fuzzy reasoning is proposed by comprehensively considering the factors that affect the ship domain.By analyzing the research results of predecessors and selecting the key factors that affect the size of the ship domain,the complexity of the research is reduced.The method of fuzzy reasoning introduces as many influencing factors as possible into the calculation of the ship domain,and constructs a ship domain that can change according to different environments and different ship properties.Secondly,based on this,a ship collision risk model based on DCPA,TCPA,D,β,and K is established,and it is compared with the previous model.The results show that the collision risk trend of the two models is generally the same,and the error is small,which proves the rationality of the model proposed in this thesis.Then,the selection operator and genetic operator of the genetic algorithm are optimized.In terms of the selection operator,a new proportional selection method is proposed,which not only ensures that excellent individuals can be inherited to the offspring but also ensures the diversity of the population.In terms of the genetic operator,the coefficients of the operator are optimized,and the target of adapting the smaller fitness to the larger coefficient and the larger fitness to the smaller coefficient is achieved.Compared with the traditional genetic algorithm,the stability and optimization speed of the optimized genetic algorithm have been improved.Finally,with the help of the Matlab platform,a collision avoidance simulation interface is designed,and simulation simulations of ship collision avoidance paths are conducted under different encounter situations.The collision avoidance paths planned not only meet the safety requirements but also meet the requirements of ship collision avoidance rules and have good economic characteristics.