Traffic jam tracing method based on multi-source data

Abstract

The invention provides a traffic jam tracing method based on multi-source data. The method comprises a map matching algorithm (AVI-MM) based on the multi-source data and a multi-level traffic jam tracing algorithm. The AVI-MM algorithm fuses sparse AVI data and massive GPS data, firstly, a vehicle path candidate set is generated based on a random walk algorithm, then, a prior probability is determined based on a Logit path selection model with the maximum utility, a conditional probability is defined through space-time fusion analysis, and finally, sub-paths with the maximum matching probability are communicated in sequence to obtain a final full path; the multi-level traffic jam traceability algorithm is based on a map matching result, and multi-level vehicle track OD distribution of road sections, nodes and areas is calculated respectively. According to the method, high-precision matching of the quasi-full-sample vehicle track and the driving path is realized, and based on the path matching result of the quasi-full-sample AVI data, the origin-destination distribution and the path distribution of the traffic flow of the congested road section are quantitatively analyzed.

Description

technical field [0001] The invention belongs to the field of urban traffic control and management, and in particular relates to a traffic jam tracing method based on multi-source data. Background technique [0002] Tracing the source of traffic congestion refers to tracing the starting and ending positions and driving paths of vehicles outside a certain spatial range. By tracing the source of traffic flow and identifying its destination, it provides strong support for the formulation of traffic control strategies. [0003] Map matching is to calibrate the vehicle position information by expressing the road segment by segment on the digital map. First, the sample vehicle trajectory data is matched to the road network through the map matching algorithm, and then the source of traffic flow passing through the observed road segment, intersection or area can be obtained. and whereabouts, and finally realize the traceability of vehicle congestion. Therefore, map matching is an im...

AI Technical Summary

Problems solved by technology

These algorithms all show the problem of insufficient matching accuracy when facing sparse AVI data

[0005] Problems existing in the existing technology: On the one hand, the existing traffic congestion traceability methods mostly use GPS data, but there are limitations such as insufficient data sample size and location deviation, and it is impossible to realize traffic congestion traceability based on large-scale samples

On the other hand, although a quasi-full sample of traffic congestion can be traced by matching massive AVI trajectory data to the traffic network, there is a lack of map matching algorithms for sparse AVI data.

Images