Research On Decision-making For Intelligent Dynamic Collision Avoidance Of Ships In Open Water

With "Made in China 2025" plan proposed,MIIT has actively promoted the development of smart ships.In 2016,MIIT established the “Intelligent Shipbuilding 1.0 Research and Development Special Project”.The goal of this project is to build a unified network and information platform based on modern Internet technologies,develop smart ship common technology,and completed the development of China Smart Ship 1.0 with integrated energy efficiency management,equipment operation and maintenance,autopilot decision support,and ship-shore integrated communication management.In this paper,the decision-making algorithm of intelligent ship collision avoidance in open water is the main research object,and a method of multi-ship collision avoidance route planning is proposed and verified by simulation.Firstly,this study analyzes the knowledge of collision avoidance in ships and studies the process of collision avoidance of ships in open water.Then the mathematical model of collision risk is established based on the analysis of situations of two ships meeting and safety encounter distance.The avoidance decisions of multi-ship encountering are briefly analyzed.Secondly,it compares the algorithms that can be applied to ship collision avoidance intelligence decision.The problem of ship collision avoidance is analyzed to point out the shortcomings of the traditional collision avoidance system,and propose an improved way to determine the danger of collision avoidance.And the purpose and constraints of the vessel collision avoidance problem is analyzed.Based on the characteristics of the collision avoidance process,Genetic Algorithm is selected out.Finally,the ship collision avoidance decision based on Genetic Algorithm is studied.The ship collision avoidance model based on genetic algorithm is established,the structure of the genetic factor and coding method is determined.The safety,economy,and rule compliance in the collision avoidance process are analyzed to determine the fitness function of the Genetic Algorithm.In Visual C++ 6.0 software,the ship collision avoidance system based on Genetic Algorithm is designed to verify the feasibility of the algorithm.The influence of genetic operation probability on the optimization result is studied to complete the result of genetic algorithm application further.