Traffic flow filling algorithm based on multi-source data fusion

Abstract

The invention discloses a traffic flow filling algorithm based on multi-source data fusion. Firstly, trip OD information of urban residents is obtained through mobile phone signaling data; bayonet records are utilized to obtain vehicle observation flow values of the corresponding road section of a bayonet in different periods; then a selected area is divided into N small areas, based on the road section with the vehicle observation flow values, and an optimal objective function is established to obtain OD sample expansion coefficient matrix alpha t among all the areas; then a genetic algorithmis utilized to obtain the optimal objective function to obtain the optimal solution of the alpha t; finally, the OD matrix after sample expansion is obtained based on the optimal solution of the alpha t, and the OD matrix after sample expansion is distributed to the urban road network to obtain vehicle simulation flow values of each road section and each period; for the road sections without thevehicle observation flow values, the vehicle simulation flow values are taken as the filling flow of vehicles in the period t. The traffic flow filling algorithm based on the multi-source data fusionhas the advantage that truer filling flow data with concurrent precision and breadth can be obtained.

Description

technical field [0001] The invention relates to a traffic flow filling algorithm based on multi-source data fusion. Background technique [0002] In the analysis of urban road traffic conditions, road traffic flow is an important indicator and the basis for accurate follow-up analysis, control and guidance. Traditional urban road traffic flow acquisition methods are mainly realized by loop coil detectors or video detectors, which have the advantage of high accuracy, but due to high input costs, the amount of data acquired is small, and in urban road networks Many roads cannot be covered. Mobile phone signaling data is a kind of activity data of urban residents with a relatively wide coverage. It has the advantage of high breadth. However, because mobile phone signaling data is the data generated by urban residents' activities, it cannot be directly filled as traffic flow data. . In urban road traffic research, there are certain requirements for the accuracy and breadth of...

AI Technical Summary

Problems solved by technology

Traditional urban road traffic flow acquisition methods are mainly realized by loop coil detectors or video detectors, which have the advantage of high accuracy, but due to high input costs, the amount of data acquired is small, and in urban road networks Many roads cannot be covered

Mobile phone signaling data is a kind of activity data of urban residents with a relatively wide coverage. It has the advantage of high breadth. However, because mobile phone signaling data is the data generated by urban residents' activities, it cannot be directly filled as traffic flow data.

In urban road traffic research, there are certain requirements for the accuracy and breadth of traffic flow distribution. The accuracy and breadth of traffic flow data affect the results of urban traffic decision-making. However, through the expression of the above data characteristics, it can be found that the advantages of a single data Some are limited, and are also limited by input costs, hardware facilities, etc.

[0003] To sum up, there are still deficiencies in the current acquisition methods of urban traffic flow, or the data coverage is small, and it is difficult to obtain data with a high breadth; or the noise and other components in the data are too high, and it is difficult to obtain high-precision data

Images