A Traffic Speed ​​Prediction Method Based on Spatiotemporal Graph Convolution-Generative Adversarial Network

Abstract

A traffic speed prediction method based on spatio-temporal graph convolution-generative confrontation network. First, according to the urban road network structure, the road sections in the road network are used as the nodes of the graph, and the intersections are used as the edges connecting the nodes to construct a traffic graph network; then , through the traffic speed detector in the urban road network to obtain sampling and collect traffic speed data, construct a feature matrix, and then, use the space-time graph convolution network (STGCN) constructed in the present invention as the generator of the generation confrontation network and the fully connected neural network The network acts as a discriminator to generate traffic speed data against each other, and through mutual game training and joint improvement, an optimal traffic speed prediction model is finally obtained. The final generator of the generative confrontation network is used to generate the traffic state data prediction value closest to the real data, so as to achieve the purpose of road network state data prediction in the present invention. The invention can realize more accurate traffic speed prediction.

Description

technical field [0001] The invention relates to the field of intelligent traffic engineering, in particular to a traffic state prediction method of an urban road network. Background technique [0002] With the continuous improvement of the national economy and the continuous advancement of urbanization, the number of urban residents continues to increase. With the continuous improvement of residents' living standards, the number of cars is also maintaining a rapid growth, and the urban traffic network has become more and more complicated, and the occurrence of transportation problems has become more frequent and complex. In order to alleviate the slow and congested traffic conditions in cities, researchers in the field of transportation begin to predict traffic data, and further improve urban traffic conditions based on the predicted data. [0003] Because traffic speed is highly time-varying, short-term traffic speed prediction is usually used for data analysis. The predi...

AI Technical Summary

Problems solved by technology

Among them, the spatio-temporal graph convolutional network (STGCN) analyzes the temporal traffic data and the graph space node neighborhood at the same time, but these prediction methods often only focus on the second-order neighborhood of the road network nodes, ignoring the global characteristics of the traffic road network

Images