Floating car track data map matching method and system

Abstract

The invention belongs to the technical field of road network matching, and particularly discloses a floating car trajectory data map matching method and system, and the method comprises the steps: carrying out the preprocessing of vehicle positioning trajectory point data, cleaning abnormal data points in the trajectory point data, carrying out the batch vertical projection of the preprocessed trajectory point data facing an adjacent road section, obtaining alternative point positions, and carrying out the positioning of the vehicle positioning trajectory point data; calculating the distance from the preprocessed track point sequence to an alternative point location, obtaining a local road network according to the alternative point location, carrying out topological processing, calculating the shortest path distance, calculating the alternative probability of the alternative point according to the distance between the alternative point location and the original track point and the shortest path distance relation between the alternative points corresponding to the front and back track points, and obtaining the alternative probability of the alternative point according to the alternative probability. And calculating a finally matched real path through the alternative probability of the alternative point. By adopting the technical scheme, two map matching technical routes of geometry and topological relation are fused, and the matching accuracy and effectiveness are improved.

Description

technical field [0001] The invention belongs to the technical field of road network matching, and relates to a method and system for matching a track data map of a floating car. Background technique [0002] Urban taxis are an important part of the main body of urban road traffic operations. Urban taxis mainly provide transportation services by cruising to obtain passengers. Therefore, taxis are also called floating vehicles. An important source of data for transportation systems. The existing vehicle-mounted GPS equipment itself has an error of 5-10 meters. When driving to an area with dense buildings and a large development volume, the error will increase. The data of floating cars with positioning deviations do not match the roads, and the data of floating cars with positioning deviations are difficult to be directly applied to the calculation and evaluation of urban road operation status. Therefore, it is necessary to solve the problem of deviations in taxi operation tr...

AI Technical Summary

Problems solved by technology

The existing hidden Markov model transition probability matrix only considers the factors of the straight-line distance between the shortest path after matching and the actual point-to-point, and ignores the factors of the distance between matching points and the actual point-to-point distance. , it is very prone to mismatching, and finally the Markov model selects the global road network for search in the path search strategy, which is less efficient

[0007] Map matching based on artificial intelligence theory requires a large number of training data sets for point-by-point matching, and a large amount of training data is required for road network matching, which is costly

Images