Research On Path Planning And Obstacle Avoidance Based On Unmanned Surface Vehicles

In recent years,with the continuous expansion of the ocean,how to explore the ocean and develop marine resources has become the focus of various countries.In this context,unmanned surface vehicles are receiving unprecedented attention,how to design a reasonable path navigation for the unmanned surface vehicles,dealing with the avoidance problem for the emergency situations has become a core issue in the research of unmanned surface vehicles.Based on the actual needs,this paper puts forward a hybrid path planning algorithm,combining with the global and local path planning to deal with the unknown environment or moving obstacles.The hybrid algorithm uses the improved particle swarm optimization algorithm to plan a global path between the beginning and the end of the known environmental;based on this path,this paper uses the improved artificial potential field method to avoid the obstacle.The unmanned surface vehicles can resume previous route after avoiding the obstacles.When dealing with global path planning problem,this paper firstly carries on environmental modeling,making the two-dimensional environment model represented by grid model;then uses the decreasing nonlinear inertia weight particle swarm algorithm to plan a shortest and the smoothest global path.For the local planning of hybrid algorithm,an improved artificial potential field method is presented.The relative velocity potential field and relative acceleration potential field are introduced to solve the dynamic obstacle avoidance ability of unmanned surface vehicles.In addition,the algorithm introduces the line potential field from the global path to make the USV sail along the planned path after avoiding obstacles.It fits the global path planning perfectly.This paper makes the hybrid algorithm applied to the electronic chart.The simulation experiment is done in the electronic chart.The experimental results meet the requirements of path planning and obstacle avoidance for unmanned surface vehicles,and verify the feasibility of the algorithm.