Font Size: a A A

Research And Implementation Of Ship Collision Avoidance Aided Decision System Based On Improved Genetic Algorithm

Posted on:2019-11-17Degree:MasterType:Thesis
Country:ChinaCandidate:Y Y LiuFull Text:PDF
GTID:2392330596465389Subject:Electronic Science and Technology
Abstract/Summary:PDF Full Text Request
With the development of the global economy,economic exchanges around the world are becoming more and more frequent,and frequent economic exchanges need high frequency transport support.When the transportation volume is relatively large,the transport modes between different continents are mainly sea transport.Therefore,the maritime transport is becoming increasingly busy,and the density of ships on routes is increasing,which increases the probability of collisions between ships and the demand for automatic collision avoidance systems for ships.In view of this background,this paper combines the improved genetic algorithm and classifies the cases of multiple ships,gives the strategy of avoiding the collision,and analyses the collision avoidance strategy through simulation experiments,and proves the feasibility and superiority of the avoidance strategy given by the ship collision avoidance decision system completed in this paper.The main research work completed in this paper is as follows:(1)The whole process of ship avoidance is analyzed.According to the motion parameters of the ship and the rules of collision avoidance at sea as the constraint conditions,the parameters such as the time of avoidance,the angle of avoidance,the opportunity of the return flight and the angle of the return navigation are analyzed and quantified.(2)An improved genetic algorithm is proposed.Aiming at the shortcoming of traditional genetic algorithm being premature convergence to local extreme value,the corresponding improvement method is given.Through multigroup evolution,each population searches for a solution space,which makes it easier to converge to the global extremum.The crossover process is improved.In one crossover operation,the direction of multiple changes is repeated many times to increase the accuracy of the solution.The Metropolis criterion is adopted to accept the new individuals and accelerate the convergence speed without breaking the population diversity.In the end,by comparison,it is proved that the improved algorithm has a stronger global optimization ability,avoiding the occurrence of precocious situation,and the optimal solution has higher precision.(3)A multi-objective optimization weight determination operator is proposed,The operator gives the adaptive weights according to the mathematical expectation relationship among the objective functions.The weight coefficient determined by the operator makes the algorithm continuously adjust the search direction in the process of evolution,taking into account the various sub-goals in the objective function,and proceeding toward multiple targets at the same time.Finally,by comparing with other documents,the rationality and superiority of the operator are proved.(4)The multiple ship encounter situation is classified and the multiple ship avoidance rules are put forward.In view of the fact that the international maritime collision avoidance rules are not specific to the avoidance method in the case of multiple ship meeting,through the analysis,the situation of multiple ships is classified and the rules of avoidance for each situation are given.(5)An assistant decision-making system for ship collision avoidance based on improved genetic algorithm is designed and implemented.In the process of collision avoidance,the nearest distance between the ships is a safety measure,the time required for the whole process of avoidance is economic consideration,and the two aspects are normalized,the target function is obtained,and a reasonable collision avoidance strategy is given by improving the genetic algorithm.
Keywords/Search Tags:analysis of collision avoidance process, Improved genetic algorithm, Collision avoidance decision making, multi ship meeting, objective function
PDF Full Text Request
Related items