Missing traffic data restoration method for complex urban traffic network

Abstract

The invention provides a missing traffic data restoration method for a complex urban traffic network. The method comprises four steps of obtaining a data source, road network matrixing, constructing atraining model, and data restoration. Firstly, the data source is obtained, comprising: performing statistics on bayonet traffic and extracting vehicle trajectory data. Then, in order to extract local space features of the road network traffic, an intersection relationship is mined from a vehicle trajectory through a network2matrix road network matrixing algorithm to reconstruct the road networkto obtain a road network traffic picture. Local convolution calculation is performed by using a convolutional neural network. Secondly, in order to obtain the time-space characteristics of the road network traffic, a multi-modal neural network-traffic flow recursive restoration neural network comprising a predictive sub-network and a restoration sub-network is constructed, and the restored data are used as inputs to perform iterative training. Finally, the road network traffic with missing data is input into the trained neural network in a time sequence for restoration to obtain a missing datarestoration result. By adoption of the missing traffic data restoration method provided by the invention, the restoration accuracy is improved.

Description

technical field [0001] The invention belongs to the fields of data mining and intelligent transportation, relates to missing data restoration, and is a method for restoring missing traffic flow data facing complex urban traffic networks. Background technique [0002] With the rapid development of traffic informatization, traffic data has grown explosively. Among them, the number of passing vehicles per unit time is an important index to describe the traffic flow, and its changing law reflects the basic nature of the traffic flow. However, due to failures in collection, transmission or storage, there will be errors and loss of data. The lack of data not only reduces the validity of the data, but also brings troubles to the follow-up work. The recovery of incomplete traffic data is an effective processing of massive traffic data, which is of great significance to the follow-up analysis of traffic management system (TMS) and intelligent transportation system (ITS). [0003] ...

AI Technical Summary

Problems solved by technology

Due to the large amount of calculation, the above-mentioned existing technologies are not aimed at complex urban road traffic; at the same time, because the local spatio-temporal characteristics of urban road network traffic data are not utilized, the restoration effect does not have good generalization ability

Images