Path Planning For Unmanned Surface Vessel

With the development of unmanned technology and robot technology,path planning technology,which is an important part in the working process of automated equipment,has become increasingly critical.Unlike the usual robot path planning,due to the influence of environmental factors such as sea conditions and weather,and the limitation that the unmanned surface vessel cannot be accurately controlled on the water surface,the path planning of the unmanned surface vessel on the water surface needs to consider whether the equipment can be safely driven in such an environment.Different algorithm strategies are used according to different tasks to ensure that the tasks can be completed safely and effectively.In actual scenarios,the distribution information of surface targets is usually unknown,so it is necessary to collect surface data through reconnaissance tasks to obtain the distribution information of target points in the area;when the distribution information is known,it is necessary to designate several unmanned surface vessels to patrol the targets to achieve the goal of continuously updating the target data.Therefore,this paper studies the planning strategies of reconnaissance tasks and patrol tasks for the path planning of unmanned surface vessels on the surface.In the reconnaissance mission,in order to ensure that the target point is not missed while avoiding the unmanned surface vessel entering the threat zone of the target point,this paper adopts a two-stage planning strategy of global pre-planning and local re-planning to realize the reconnaissance of the unmanned surface vessel to the target point.Global pre-planning ensures the complete coverage of the mission area by the sensors carried by the unmanned surface vessel;local re-planning ensures the safety of the unmanned surface vessel during the short-range reconnaissance of the target point,and prevent the unmanned surface vessel from entering the threat zone of the target point.Because local re-planning will modificate the path,the coverage formed in the pre-planning stage might be destructed.To avoid this problem happened,this paper uses a zoning strategy throughout the path planning process to ensure that the re-planned path can still form a complete coverage of the task area.Through analysis and experiment verification,this method can effectively detect all target points in the mission area,and achieve the purpose of complete reconnaissance.In the patrol task,considering the cost of surface unmanned surface vessels,this paper assigns a time penalty factor to each target point under the condition of limiting the number of sensors,proposes a minimum penalty scan coverage problem,and gives an equal-split strategy and branch and bound Algorithmic strategies and solutions based on greedy strategies.Due to the high time complexity of the branch and bound strategy,it is not suitable for the case of large-scale target points.In this paper,experiments of different scales are set up to compare the branch and bound algorithm,the greedy algorithm,and the most basic equal-split algorithm.Through complexity analysis,it can be proved that the time complexity of the greedy algorithm is 2O(| V |),the best case time complexity of the branch and bound algorithm is 2O(| V |),and the worst case time complexity is |V| 2O(| V |?2-| V |);in the experiment,the penalty value of the greedy algorithm is only 3.26% greater than that of the branch and bound algorithm,and it is better than the equal-split strategy on small-scale data and large-scale data.