The Research On Path Planning And Obstacle Avoidance Methods For Unmanned Surface Vehicle

The research and construction of high-level,modern and intelligent unmanned surface vehicle(USV)are of great significance to our country's strategy of achieving a maritime power.The path planning and obstacle avoidance technology is a key step for autonomous navigation and operation of USV,which represents the intelligent level of USV.There are many defects for the current path planing,such as unsmooth paths,easy to fall into local optimal paths,and high complexity of collision avoidance,which lead to problems such as large corners,long routes,and even obstacle avoidance failure of planned paths for USV.Aiming at these problems,this paper conducts theoretical research on global path planning and local obstacle avoidance of USV,and carried out relevant experiments of USV in combination with reality.First of all,combining with the characteristics of the USV navigation environment,this paper uses particle swarm optimization to study the global path planning.In order to solve the problem that the particle swarm algorithm is easy to fall into local optimality,the chaos theory is integrated into the basic particle swarm optimization algorithm.Through chaotic iteration of the current global optimal value,a chaotic population is generated and some particles that are partially local optimal are replaced,which improves the problem of insufficient particle diversity in the late stage of population search;at the same time,in order to strengthen the local search ability of the algorithm,the article combines the particle swarm algorithm with the following bee search strategy of the bee colony search algorithm,and proposes the chaotic particle swarm-bee colony algorithm.Combining the proposed algorithm with cubic spline interpolation,and designing population fitness evaluation function with penalty function to solve the unmanned boat global path planning problem.The simulation verifies the advantages of the search algorithm in terms of convergence speed and search accuracy.Secondly,in order to realize the collision avoidance of local obstacles in the course of USV navigation,a local collision avoidance method based on particle swarm optimization algorithm and maritime collision avoidance rules is adopted in this paper.This method introduces fuzzy mathematics theory to judge the collision risk of USV and obstacles,and establishes a collision avoidance model which transforms the problem of local obstacle avoidance into the optimization of the speed and course variation of USV.What is more,this paper integrates the maritime rules and the USV dynamics constraints further processed the range of speed and course,A group of optimal obstacle avoidance strategies are calculated by iterative optimization of particle swarm optimization.The typical encounter situation of USV and maritime vessels and the complex marine environment were selected for simulation study,which verified that the proposed method could be well adapted to the situation of unmanned boat avoiding local unknown obstacles.Finally,the path planning and obstacle avoidance methods were verified on the working platform of the unmanned boat.The offline global path planning experiment proved that the improved algorithm combined with cubic spline interpolation has the advantages of short planning path and good path smoothness.The local obstacle avoidance test based on this paper realizes the safe navigation of the USV and verifies the feasibility of the local obstacle avoidance method in this paper.