Traffic flow completion and prediction method

Abstract

The invention discloses a road traffic flow prediction method, and the method comprises the steps: sampling the historical traffic information of a road condition, and constructing a traffic network model; providing a missing data complementation method based on historical traffic information, building a reachable matrix according to travel time, carrying out the graph convolution through employing the reachable matrix as a convolution kernel, and extracting features; training a recurrent neural network according to the extracted features and the collected traffic data to obtain a flow prediction model; based on real-time sampling traffic data, inputting a pre-trained prediction model, obtaining a new error result in the process, and dynamically training the model.The reachable matrix canreduce unnecessary space search, the space-time relationship mining efficiency is improved, and the road traffic flow prediction accuracy under the conditions of more missing of road traffic flow dataand longer prediction duration is improved.

Description

technical field [0001] The present invention relates to the field of traffic technology, in particular to a road traffic flow prediction method based on deep learning and graph network technology. Background technique [0002] In recent years, with the continuous development of technology, urban traffic pressure is increasing day by day. Timely and accurate forecasting of traffic flow plays a key role in planning traffic routes, improving travel efficiency, and relieving traffic pressure. [0003] In the existing road traffic flow forecasting models, the missing data is usually only filled with zeros. When predicting, only the time dependence between the traffic flows is often considered, and the spatial dependence between them is ignored. Therefore, it cannot To achieve real-time and accurate forecasting effect. [0004] In real life, the traffic conditions of the current road will obviously be affected by the traffic conditions of adjacent roads. However, because the tr...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome factors such as the single feature in the existing road traffic flow prediction model and the inability to effectively consider the traffic conditions of adjacent roads, resulting in low accuracy of the model, the present inve

Images