Multi-ship Collision Avoidance Strategy Based On Particle Swarm Optimization Algorithm

As the marine traffic environment is more and more complex and the ships are sailing more and more densely,the ship collision accidents occur from time to time.Therefore,it is urgent to strengthen the research of multi-ship intelligent collision avoidance strategy.At present,the research on multi-ship collision avoidance strategy seldom considers the constraints of navigation rules and actual navigation habits,and the collision avoidance strategy is not efficient and safe.Therefore,in this paper,against the background of multi-ship encounters in wide sea area,navigation rules are integrated into the improved particle swarm optimization path planning algorithm,and a multi-ship collision avoidance strategy is designed based on the combination of global path planning algorithm and deterministic algorithm.Several major problems of multi-ship collision avoidance strategy were solved,which are inconsistent with actual navigation rules and habits,low collision avoidance efficiency and low safety.The main research contents of this paper are as follows:firstly,the problem of collision avoidance of multiple ships is analyzed.The differences of multi-ship encounter between wide and complex waters are studied,and the conclusion that multi-ship collision avoidance in open waters is decomposed into three ship encounter is put forward,and the research scope is further determined.The representation method of ship collision hazard and the common safety field of ship collision avoidance were studied,and determines the safety field and hazard model.The advantages and disadvantages of particle swarm optimization algorithm and genetic algorithm are analyzed and compared.Secondly,aiming at the local optimal value of particle swarm optimization algorithm and the inconsistency between the planned path and the actual navigation path in multi-ship collision avoidance applications,the improvement strategy of particle swarm optimization algorithm is analyzed.Adaptive and heuristic methods are used to improve the particle swarm optimization algorithm.The adaptive ability of particles is enhanced from the two aspects of particle evolution speed and particle aggregation degree.Navigation rules,route safety and economy are added into the fitness function of particle swarm optimization path planning algorithm,and a rule adaptability judgment method based on trajectory slope change is designed.The problem of inconsistency with the actual voyage was solved.Finally,the collision avoidance strategy of multiple ships is studied.A step-by-step collision avoidance strategy based on particle swarm optimization path planning algorithm and deterministic algorithm is designed.In the collision avoidance strategy,a hazard judgment method based,on the ship's navigation history is designed to judge the obstacle ship that does not conform to the navigation rules.Simulation in MATLAB environment and compared with the collision avoidance strategy of the real-time online path planning,the results show that the collision avoidance strategy combining the path planning algorithm with the deterministic algorithm can improve the collision avoidance efficiency of the collision avoidance strategy to a certain extent,and is more in line with the requirements of navigation rules and navigation habits,and has higher navigation safety.