Research On Ship Automatic Collision Avoidance Decision Based On Improved Particle Swarm Optimization Algorithm

Ship collision accidents will not only cause significant economic losses,but also engender pollution of the marine environment and serious harm to people’s health.Human factors are the primary factor for ship collision accidents.Therefore,this article focuses on ships automatic collision avoidance decision-making in open waters.The main research results of this article are as follows:(1)Summarize the basic theory of ship collision avoidance in detail,and divide the situation of ship encounters combined with the "International Regulations for Preventing Collisions at Sea".The determination of key avoiding ships,the calculation of ship collision avoidance parameters,and the calculation of ship collision risk are also explained in detail.A method for calculating the latest rudder point based on the ship’s trajectory is proposed in this article.This method takes into account ship’s minimum safe passing distance,steering response time,ship speeds,encounter situation,turning time,advance distance,and transverse distance.The decrease of ship speed during the turning process is also considered,so it is more consistent for practical navigation.(2)Introduce the basic principles and improvement strategies of particle swarm optimization.Aiming at the problems of particle swarm optimization easily falling into local optimality,premature convergence and slow later iteration speed,combined with the crossover operator in genetic algorithm,and adopt the adaptive weighting method to improve particle swarm algorithm.A new type of hybrid particle swarm optimization algorithm based on the global optimal point to dynamically adjust the inertia weight is proposed in this article.(3)Construct objective functions based on safety,economy,collision avoidance rules and timing of avoidance.Due to the contradiction among the objective functions,obtaining the optimal solution of each objective function is difficult at the same time.Therefore,this paper adopts the linear weighting method in multi-objective optimization method by giving each objective function a certain weight value,and constructs the total objective function of ship collision avoidance decision,then use the improved particle swarm algorithm to solve the total objective function.(4)In this article,Matlab2018a is used as the simulation platform.In the open sea,using standard particle swarm algorithm and the improved particle swarm algorithm to conduct algorithm comparison simulation experiments on the collision avoidance problem of two ships in three classic encounter situations: encounter,crossing and chasing.and it is concluded that the optimization effect of improved particle swarm is better.Finally,using the improved algorithm to simulate the three-ship encounter situation and the four-ship encounter situation.According to the simulation results,the hybrid particle swarm optimization algorithm based on the global optimal point to dynamically adjust the inertia weight can effectively solve the automatic collision avoidance in the situation of multi-ship encounters.