Path Planning Method Of Unmanned Surface Vehicle Based On Probabilistic Roadmap

As an intelligent Unmanned platform,Unmanned Surface Vehicle(USV)is an important product of the development of Marine technology.The development of USV technology revolves around the mission content,and it has been widely used in the field of Marine exploration and development and military applications.The autonomous navigation technology of USV has become one of the main problems,and the path planning problem has become a key link.Efficient,reliable and real-time obstacle avoidance path planning is crucial for USV.Therefore,on the basis of understanding the common path planning methods and summarizing the existing problems of USV path planning,this paper proposes a path planning algorithm based on probabilistic map sampling.According to the function of USV path planning,the common planning methods are classified and compared,and the characteristics of USV path planning are summarized.Through the understanding of the status of different environmental modeling methods,the problems of USV path planning algorithms are summarized from the perspective of different environmental models and sampling search methods.Based on the basic principle of the prototype of probabilistic map method,this paper improves it.The grid method is used to divide the original map into feasible and infeasible areas.When constructing the map,the edge expansion is used to represent the high tide and low tide.Random sampling is used to obtain a number of points that are located at the vertices of the grid and do not coincide with obstacles.According to the RRT algorithm,the method of randomly expanding the tree to generate new nodes is used to avoid the planning failure task caused by the imbalance of planning space and step size,which takes a lot of computing time.The improved PRM method can reduce the amount of calculation by limiting the number of sampling points,so as to ensure that the path search can obtain the optimal path or suboptimal path more quickly,and achieve better results by adjusting the number of nodes and limiting the step size.Secondly,this paper summarizes the collision avoidance process for maritime navigation.The obstacle identification and modeling method were expounded,the collision avoidance motion parameters between USV and obstacles and the minimum encounter distance and time after adjusting the speed were calculated,and the calculation method of collision risk was introduced.The possible encounter situation between USV and obstacles is analyzed,and the collision avoidance measures should be made by both sides in accordance with the provisions of international maritime regulations.The artificial potential field method is fused and the virtual potential field is introduced to solve the problem of local obstacle avoidance.Finally,its effectiveness was verified by global path planning simulation of different tidal environment Settings and grid map established by electronic chart.Secondly,the influence of the number of sampling points and step size restrictions on the results of global path planning was summarized by comparing the results and time of different control groups,and the local path planning with virtual potential field was simulated and verified.The local obstacle avoidance of the path between two nodes is completed by the superposition of the gravitational potential field and the repulsive potential field.