Vehicle trajectory reconstruction method based on license plate recognition data

Abstract

The invention discloses a vehicle trajectory reconstruction method based on license plate recognition data, and the method comprises the following steps: 1), collecting urban road network topologicalinformation and information, i.e., license plate recognition data, that individual vehicles on a road network pass through a road gate in a certain time period, and preprocessing original data; 2), interrupting a travel chain according to reasonable travel time acquired by an online map, and separating a complete path set from an incomplete path set; 3), generating a candidate path set in combination with a space-time prism theory and a K shortest path algorithm; 4), formulating evaluation indexes for the candidate paths and determining an index normalization method; and 5), with an automaticencoder, carrying out decision making to realize restoration of incomplete vehicle trajectories. According to the method, restoration of the incomplete vehicle trajectory caused by missing detection in an urban road network can be completed with high precision, the algorithm speed is high, the robustness is good, and the method has good performance in an actual traffic scene.

Description

technical field [0001] The invention relates to a vehicle trajectory reconstruction method based on license plate recognition data, and belongs to the technical field of vehicle trajectory reconstruction in intelligent transportation. Background technique [0002] Motor vehicle travel paths can not only reflect the characteristics of traffic flow from a microscopic perspective, but also provide data support for the analysis of road network traffic demand and spatio-temporal distribution characteristics. Extracting and analyzing the travel paths of all running vehicles in the road network can restore the actual operating state of the road network. However, because the traffic information collection equipment cannot fully cover the road network and vehicles, and the equipment itself will malfunction and cause data packet loss and other problems, part of the obtained vehicle travel trajectories is incomplete. Therefore, reconstruction of these incomplete trajectories is requir...

AI Technical Summary

Problems solved by technology

In the existing candidate path set generation technology, most of them use the K-shortest path algorithm, but this method is slow in searching large-scale road networks, and there may be problems that the candidate paths do not conform to the actual situation

At the same time, in the existing route decision-making technology, most of the decision-making indicators are static, which cannot reflect the different perceptions of travelers on the same route. At the same time, the determination of the weight is subjective to a certain extent, and there are many calculation steps.

Images