Collision Avoidance Decision Making Method For Multi-target Ships At Sea

In recent years,computer technology and information technology have developed rapidly,and modern ships are advancing at a high speed in the direction of automation and intelligence.More and more ship researchers are working on marine automation research in order to liberate ship drivers and avoid the impact of navigation on human safety.Therefore,according to the actual needs,this paper designs a ship collision avoidance decision-making subsystem,embedded in a large ship navigation equipment software,in order to provide collision avoidance decision when there is a collision risk with other ships or obstacles.The main research contents of this paper are as follows:1.Analyze the current situation of intelligent collision avoidance of ships at home and abroad,and master the calculation method of ship collision avoidance principle and related motion parameters.The situation of the encounter between ships is divided in detail.The calculation model of ship collision comprehensive risk is given from the perspective of time and space,and the process and basic requirements of ship collision avoidance are analyzed.2.Fully study the basic principles and characteristics of genetic algorithms,and analyze and improve the problems existing in genetic algorithms.The purpose of improving the performance of the genetic algorithm is achieved by optimizing the three basic genetic operators of selection,crossover,and mutation.The performance of the improved genetic algorithm is compared with the traditional genetic algorithm through MATLAB.The comparison results show that the improved genetic algorithm is superior to the traditional genetic algorithm in convergence,stability and anti-early maturity.3.Considering the influence of ocean current on ship navigation,and combining the improved genetic algorithm with the ship collision avoidance model,the fitness function is designed by using floating point number coding to encode each avoidance route.And established a ship collision avoidance decision-making subsystem.When the fitness function is given,not only the economics and safety of the route are considered,but also the constraints of the International Collision Avoidance Regulations on the collision avoidance process of ships are introduced.At the same time,the problem of collision avoidance for ships is given.Performance evaluation criteria.4.Embed the ship collision avoidance decision-making subsystem into the navigation device software,connect the simulator to the real-time navigation information and the target ship information to simulate the actual encounter situation between the ships.The actual simulation results show that the ship collision avoidance decision-making subsystem based on the improved genetic algorithm can plan a collision avoidance route when encountering the collision risk,and evaluate the collision avoidance route through MATLAB simulation results to verify the effectiveness of the collision avoidance result.The simulation results show that the collision avoidance path planned by the ship collision avoidance decision-making subsystem is basically correct,the execution efficiency is high,and the decision-making scheme can be provided quickly.At the end of the paper,the shortcomings of this topic are pointed out,as well as the subsequent research directions and work priorities.