Research On Complete Coverage Path Planning Of Unmanned Surface Vehicle

With the exploration of deep sea area,the protection of water resources and the development of marine military,the research and development of unmanned surface vehicle has been paid more and more attention.Among them,path planning is the key technology for unmanned surface vehicle to realize unmanned,intelligent and autonomous control.Complete traversal path planning is a special form of unmanned surface vehicle path planning,which has important theoretical significance and practical application value.It can be applied to many aspects,such as surface floating object cleaning,bottom topography detection,gulf water clearance and so on.Constructing a suitable environmental model is the first step in unmanned surface vehicle path planning.In order to improve the accuracy of outdoor water environment modeling,the traditional autonomous learning environment modeling method is optimized by expanding the boundary.According to the longitude,latitude and heading angle collected by unmanned surface vehicle,the boundary longitude and latitude data are obtained.The environment modeling is completed based on grid method.In order to solve the problems of time-consuming,laborious and tedious process of unmanned surface vehicle passing through autonomous learning environmental boundary method,as well as the low applicability and narrow application range of the modeling method based on electronic chart,an environmental modeling method for human-computer interactive publishing task is realized.In this method,the user publishes the execution task area through the upper computer software electronic map interface of the ground station,and converts the task area into a raster map.The experimental results show that the proposed algorithm is effective and feasible,because the error can be controlled within 2.61% of the actual distance.In order to solve the problems of low efficiency and poor universality of unmanned surface vehicle completes traversing path planning algorithm in complex environment,this thesis proposes a helix complete traversing algorithm based on improved A-Star algorithm.The algorithm starts traverse through the inner spiral traversing algorithm.When the unmanned surface vehicle encounters the "dead angle",the water wave algorithm is used to search the nearest untraversed grid,and then the improved A-Star algorithm is used to jump out of the "dead angle" and continue traversing.The experimental results show that compared with the existing fully ergodic path planning algorithms,thealgorithm improves the performance of the algorithm,and the number of path steps can be reduced from 814 to 784,the repetition rate can be reduces from 7.8% to 3.98%.The effectiveness of the proposed method is verified by building an unmanned surface vehicle experimental platform.The experimental results show that the environmental modeling of the school lake is completed by expanding the boundary mode,and the error is controlled within 1.6 meters.The unmanned surface vehicle completes traversing algorithm base on the improved A-Star algorithm achieves 90%coverage of the school lake.