Data-driven taxi route recommendation method

Abstract

The invention discloses a data-driven taxi route recommendation method, which comprises the following steps of: firstly, extracting an area in which an optimal route is most likely to appear through a starting point position and a final point position, then, removing network nodes which are most likely to be incapable of forming the optimal route by using an artificial potential field method, and finally, finishing optimal route recommendation by using a Dijksta method. Simulation experiments show that compared with an ant colony algorithm, a greedy algorithm, an artificial potential field method, a fast exploration random tree and a Dijkstra algorithm, the method has higher convergence efficiency and higher calculation speed.

Description

technical field [0001] The invention relates to the technical field of route recommendation, in particular to a data-driven taxi route recommendation method. Background technique [0002] In recent years, with the continuous progress of the society, people's consumption power is increasing, the frequency of travel is getting higher and higher, and the demand for urban transportation is increasing. Taxi is convenient, safe, fast, and highly personalized. More travelers choose. The increase of urban transportation has led to the increase of road congestion. For the urban congestion problem, city managers have invested a lot of money to expand and renovate urban roads, making the transportation network more and more complicated. Although the congestion has been alleviated, the urban congestion problem Still unresolved, at the same time, the increase in the road network and the congestion of road conditions make taxis less efficient in carrying passengers. At the same time, res...

AI Technical Summary

Problems solved by technology

[0007] However, some of the above existing algorithms use the idea of ​​improving the classic path planning method to achieve optimal path planning. The algorithm has high execution efficiency, but it is difficult to obtain the optimal path; the other part focuses on using hybrid algorithms for optimal path planning. The planning ability is strong, but the execution efficiency of the algorithm is low, and it is difficult to meet the requirements of fast path planning; in addition, although some of the above studies use Dijkstra's improved method for path planning, their focus is on solving large-scale complex networks. The effect is not necessarily optimal under the scale of nodes

Images