Research On Route Design Of Ship Weather Routing Based On A* Algorithm

As one of the most important tools of transportation,ships are rapidly developing toward automation and intelligence.Reducing navigational costs and ensuring navigation safety are particularly important for ocean-going vessels because of the long sailing time and the long sailing distance.Highquality ship weather routing services can ensure the safety and economic benefits of ocean-going ships.With the advent of the artificial intelligence era,the quality of ship weather routing is increasingly demanded.To complies with the development trend of ship weather routing,from the perspective of global meteorological route planning and local meteorological route planning,we take voyage time,fuel consumption,navigational risks or multiple objectives as the optimization goal,and combine with rolling window and multi-objective optimization in order to design the ship weather routes in this article.First,the preprocessing of ship weather routing.Build an environment model using square grid graph with eight moving directions.Define optimization objectives as well as establishing corresponding models for sailing time,fuel consumption and navigational risk respectively.Calculate the speed loss of the ship sailing in the wind and waves based on the empirical formula that considers the influence of the wave field and wind field simultaneously.Download the required numerical weather forecast information from the European Center for Medium-Range Weather Forecasts(ECMWF).Using the numerical weather forecast information,the simulation of ship weather routing is performed by means of a drawing toolkit--M_Map with MATLAB.Second,the study of global meteorological route planning based on improved A* algorithm.Make all concave obstacles into corresponding convex obstacles and inflate all obstacles.According to the position of the target point relative to the starting point,an improved A* algorithm is proposed by limiting the A* algorithm's search direction of each node.The simulation of global ship weather route planning shows that the improved A* algorithm can not only find the optimal path,but also can effectively reduce the number of nodes that the algorithm needs to search during operation.Compared with the classical algorithm,the improved algorithm reduces the number of search nodes by 29.25 percent.Third,the design of local meteorological route planning based on rolling window.Using the idea of rolling window,realize the dynamic planning of local ship weather routes,namely online planning.Determine the coordinates of local sub-targets.For the situation that the local sub-target falls into the static obstacle,the strategy of finding alternative points for the local sub-target is proposed by rotating the line segment connecting the current position and the initial local sub-target point alternately counterclockwise and clockwise.The A* algorithm is used to plan the local optimal route from the current node to the local sub-target.The local ship weather route planning algorithm based on rolling window and its simulation are given.The simulation results show that the algorithm can adapt to the dynamic and uncertain environment to a certain extent.It can reduce the scale of the problem,reduce the amount of on-line calculations,and make the route planning more real-time.Finally,the research of global ship weather route planning based on multiobjective A* algorithm.The navigation area is divided by square grids with eight moving directions.In order to improve the search efficiency of the algorithm,the search area of the algorithm is rectangular limited according to the latitude and longitude of the starting point and the target point.Based on the existing multi-objective A* algorithm(NAMOA*),the following improvements have been made:(a)Due to the diagonal movement,using the diagonal distance formula to design the heuristic function;(b)In order to make the algorithm faster when approaching the starting point,and more accurate when approaching the target point,the heuristic function and the weighting function that meet the Butterworth function are multiplied to achieve dynamic weighting of the algorithm;(c)After obtaining the Pareto optimal solution set,according to the user's expectations for each goal,the user is recommended to the route that is closest to the expectation,and the user's personalized route recommendation is achieved.The simulation of the global weather route planning shows that the improved algorithm can obtain more optimized and more comprehensive results,and can recommend the route closer to the user's expectations to the user,and the execution time of the algorithm is within a reasonable range.