Intelligent route recommendation method and system based on global historical ship tracks

Abstract

The invention provides an intelligent route recommendation method and system based on global historical ship tracks, and the method comprises the steps: firstly collecting AIS data, carrying out the preprocessing of the collected AIS data, obtaining ports and key nodes based on the preprocessed AIS data, constructing network nodes, dividing segments generated in the preprocessing process according to key nodes to obtain node segments, calculating the node segments based on a kernel density estimation method to obtain recommended segments, and taking the navigation distance of the recommended segments as the weight of the edge between the starting node and the ending node of the recommended segments to construct the route network; and finally, based on the constructed route network, performing calculating through a recommendation or prediction algorithm to obtain an optimal route path, and displaying a calculation result of the optimal route path to a user. According to the method and system, the global route segment network is constructed and stored by utilizing the high-quality route data screened after massive AIS data is processed, so that the planning capability of a dynamic route can be effectively improved, and a more diversified route query function is provided.

Description

technical field [0001] The invention relates to the technical fields of shipping informatization and intelligence, and in particular to an intelligent route recommendation method and system based on global ship historical tracks. Background technique [0002] Shipping route planning and optimal route recommendation are classic problems in the shipping industry. A traditional way of thinking is to rasterize and mesh the target sea area, and transform the route design problem into a general graph theory and dynamic programming problem, such as modeling the optimal route problem and using an improved dynamic programming algorithm to solve it; another A relatively novel idea is to recommend the optimal route based on the ship's historical route data, mainly AIS records. [0003] However, the above two ideas currently have problems and defects: the traditional route planning method has the problem that the model extraction cannot reflect the real route situation: the real shippi...

AI Technical Summary

Problems solved by technology

[0003] However, the above two ideas currently have problems and defects: the traditional route planning method has the problem that the model extraction cannot reflect the real route situation: the real shipping scene is affected by complex factors, and the route planning algorithm based on the mathematical model cannot have input factors. To achieve comprehensive consideration, in fact, different ship types, deadweight tons, weather, ocean currents, tides and even seasonal factors will affect the recommendation results, especially when planning int

Images