Energy-efficient Path Planning Study For Unmanned Surface Vehicle

Unmanned Surface Vehicle(USV)is a small to medium-sized watercraft that possesses autonomous planning and navigation capabilities,and is widely used in maritime search and rescue,patrol,and resource exploration.Its main systems include guidance,navigation,and control systems.Among these,the guidance system is one of the key technologies of USV,requiring the automatic planning of obstacle-free paths from the starting point to the end point based on indicators such as economy and safety,including global and local path planning,and adapting to changes in environmental information in a timely manner.Currently,the algorithms for global path planning for USV are relatively mature and complete.However,many current path planning algorithms do not consider the effects of marine environment(such as sea wind,wave and current)on USV,leading to the planned path needing to cope with adverse conditions such as headwinds and head currents.In addition,small and medium-sized USVs have high sailing speeds and low endurance in water,which makes them more susceptible to the influence of marine environment.To overcome the impact of these environmental factors on the force exerted on the vessel,more energy consumption will be generated.Therefore,planning a path for USV that takes into account the impact of the marine environment in real-time and maximizes the use of favorable sea wind and sea current to reduce energy consumption is crucial.This thesis relies on environmental data sets to build a dynamic environmental model,establishes the kinematics and dynamics models of unmanned vessels based on real ship data,and constructs an energy consumption model to improve the A* algorithm to plan paths that consider both energy consumption and sailing costs,making it more suitable for actual sailing needs.The main work content and innovative points of this thesis are as follows:(1)Construct a dynamic safe water depth model.Currently,the environment modeling methods for path planning mostly use grid methods.However,grid methods are not suitable for expressing large-scale environmental data sets,and the accuracy of the model depends more on the grid size setting,which is not conducive to safe navigation.This thesis uses Voronoi diagrams for spatial representation,with candidate paths maximizing distance from obstacles.Based on S-57 standard electronic nautical chart shoreline coordinate data and water depth point elevation data,the spatial object information is preliminarily described through Voronoi diagrams.The tidal water level information is represented as a function of time,and the actual water depth value is obtained by adding the static water depth value after kriging interpolation,followed by further construction of the dynamic safe water depth model using the local Voronoi diagram update algorithm.(2)Construct a USV energy consumption model.In order to fully consider the force exerted on the vessel by marine environment during the USV operation period,mathematical models for marine environment are respectively established and introduced into the kinematic equation of USV to solve the variation law of its speed with time under the influence of marine environment.The USV energy consumption model is constructed to analyze the energy consumed in each movement process.(3)A USV energy-saving path planning algorithm considering sailing costs.Planning algorithms based solely on energy consumption indicators will greedily consider favorable sea wind and sea current to plan paths,which may result in higher sailing costs and may not meet actual sailing needs.This study improve the A* algorithm,which considers the distance factor when planning an energy-saving path.In addition,to consider the characteristics of marine environment changing over time,this thesis plans the energy-saving path by continuously updating the path in the dynamic safe water depth environmental model according to the time breakpoints of the total mission sailing distance.Finally,a path smoothing algorithm is proposed to reduce the number of turning points and smooth turning angles.Different strategies are validated and analyzed.Firstly,the most energy-efficient paths without considering distance factors are planned separately under favorable and adverse flow conditions,and compared with the shortest path,resulting in an improvement of 43% and 5%in the ESR,respectively,with an energy-saving rate 49.9% higher under favorable flow conditions than under adverse flow conditions;Secondly,the energy-efficient paths considering distance obtained through the improved A* algorithm are compared with the shortest path,resulting in a 36% improvement in energy consumption;In addition,the energy consumption of USVs in time-varying environments is considered,resulting in a 34% increase in energysaving rate;Finally,the influence of USV speed on the energy-saving rate of the three paths is investigated,and the path considering both energy consumption and distance costs is less affected by the speed and can stably express the relationship between energy consumption and speed.The experimental results above demonstrate that the proposed energy-efficient path considering distance improves the endurance of unmanned surface vehicles while ensuring shorter voyages.