Research On Intelligent Auxiliary Decision-making Of Ship Collision Avoidance

Today,the world’s ocean shipping is still the main form of freight transportation for commercial trade.With the proposal of the state’s "21st Century Maritime Silk Road" plan,China’s foreign trade exchanges will become more and more frequent,at the same time,the number and flow of ships will increase every year,driven economic development.However,the rising labor costs and the increasing number of collision accidents have also followed.However,the rising labor costs and the increasing number of collision accidents have also followed.In order to solve this problem,the Ministry of Industry and Information Technology established the “Smart Ship 1.0 R & D Special Project” in 2016.The main goal is to use advanced information technology to realize the intelligence of the ship’s key systems,intelligent decision-making and control of ship navigation,and intelligent management of energy efficiency.So as to better ensure the safety and efficiency of the ship’s navigation.In this context,this paper conducts research and simulation verification of intelligent collision avoidance decision-making methods for ships in open waters.This article first introduces the basic knowledge in the field of ship collision avoidance and studies the basic principle in open waters.According to the 1972 International Regulations for Preventing Collisions at Sea,the meeting situation and avoidance responsibilities of ships were divided.The calculation of ship motion parameters and ship risk model are studied.Then,based on the collision avoidance geometry algorithm,two ships’ collision avoidance assistant decision-making was realized.The kinematics modeling and simulation of the ship with KT equation were carried out to facilitate the verification of the model based on the latest steering point.The theoretical calculation of the latest rudder point was carried out and visual simulation was performed to verify it.A risk model was established using the latest steering point to determine the timing of collision avoidance.The collision avoidance geometry was used to determine the avoidance range,and the avoidance strategy was formulated in accordance with the COLREGS to determine the timing of resumption.Auxiliary decision-making for collision avoidance of two ships and three ships in open waters was implemented and simulation verification was performed.Simulation results show that the algorithm can effectively achieve collision avoidance for two ships and three ships,but it is not suitable for dealing with situations where more than three ships encounter.Then,the collision avoidance path planning of the ship was realized based on the genetic algorithm.The problem of collision avoidance of ships is first converted into a general form that can be processed by genetic algorithms,and a reasonable coding method is determined.The objective function and fitness function are constructed by considering various factors such as the safety,economy,and degree of compliance with the COLREGS.Choose the appropriate genetic manipulation.The simulation results show that the collision avoidance decision-making based on genetic algorithm can realize the collision avoidance path planning for multiple ships(including three or more ships),but the algorithm has a slow convergence speed and is not suitable for real-time applications.Finally,the ship’s collision avoidance assisted decision-making was realized based on the dynamic programming idea.Firstly,a radial grid map is constructed based on the route information,and the evaluation function is constructed from the consideration of the safety of the ship,the economy of the route,and the deviation of the route.Based on this,a dynamic programming algorithm based on radial grid map is proposed.A collision avoidance simulation system was established based on Qt and VS2008.The simulation results show that the algorithm can effectively implement collision avoidance decision-making for multiple ships(including three or more ships).Finally,the bounding box algorithm is used to represent the static obstacles,and simulations verify the effectiveness of the algorithm on the static obstacles.