A Traffic Congestion Duration Prediction Method Based on Multi-source Data Feature Extraction

Abstract

This patent discloses a traffic jam duration prediction method based on multi-source data feature extraction, which includes three steps: (1) feature extraction based on multi-source data; (2) traffic jam state prediction based on road features; (3) Traffic congestion duration prediction based on feature extraction from multi-source data. This patent considers the road traffic characteristics used in the traffic congestion state prediction process, and at the same time uses the output of the traffic congestion state prediction method, that is, the predicted traffic congestion state as one of the basis for the congestion time prediction, and improves the method's measurement of the traffic congestion dissipation time. accuracy.

Description

technical field [0001] The invention relates to the field of intelligent traffic management, in particular to a traffic congestion duration prediction method based on multi-source data feature extraction. Background technique [0002] With the increase of the number of motor vehicles in my country this year, the contradiction between the transportation demand and the transportation supply capacity of the existing transportation infrastructure is becoming more and more serious, and traffic congestion is the most important contradiction between the transportation demand and the transportation supply. reflect. In the process of traffic management department's management of road traffic status, measures cannot be taken when traffic congestion has already occurred and constituted economic losses. It is necessary to predict the location and time of traffic congestion and take corresponding measures in time to reduce traffic congestion. adverse effects on the road network. Therefo...

AI Technical Summary

Problems solved by technology

At the same time, in the current domestic patents, the data used in the traffic congestion state prediction process is a single data source such as the data of floating vehicles or road fixed detectors, which leads to the prediction model used in the traffic state prediction process The obtained data structure is too simple, which may cause the model to fail to reflect the characteristics of complex traffic conditions under actual conditions.

Images