Method for generating traffic state virtual detector based on GE-GAN

Abstract

A method for generating a traffic state virtual detector based on GE-GAN comprises the following steps: 1) constructing a road detector network according to the position of a detector and an adjacentrelationship, and embedding the detector network into a low-dimensional representation vector by using DeepWalk in a graph embedding algorithm to obtain an adjacent road traffic state matrix; 2) acquiring an adjacent road traffic state data matrix, performing data normalization on the adjacent road traffic state matrix by adopting a maximum and minimum standardization algorithm, and respectively determining model structures of a generator and a discriminator to realize generative adversarial network model construction; and 3) defining a loss function of a generator and a discriminator of the generative adversarial network, taking the adjacent road traffic state data under the sliding window as input of a generative adversarial network model, and minimizing the difference between the generated data and real data distribution through adversarial training to generate the road traffic state of the virtual detector. The maintenance cost of the road traffic state detector is effectively reduced.

Description

technical field [0001] The invention relates to a method for generating a traffic state virtual detector based on a graph-embedded generative confrontation network (GE-GAN), which belongs to the field of intelligent traffic. Background technique [0002] With the rapid development of cities, the number of vehicles on the road is increasing day by day, and the problem of vehicle congestion caused by too many vehicles is becoming more and more serious. In recent years, artificial intelligence has developed rapidly and is widely used in the field of intelligent transportation, such as traffic flow prediction, congestion prediction, etc. A large amount of accurate road traffic state data is the prerequisite for intelligent transportation system to realize accurate traffic state prediction. The detection of road traffic status requires a large number of detectors, and the deployment and maintenance of road traffic status detectors require high costs. Therefore, it is of great si...

AI Technical Summary

Problems solved by technology

[0003] Traditional road traffic status data is obtained from various coil sensors arranged under the road. The detector will not only cause damage to urban roads, but also the later maintenance cost of the detector is relatively high.

The premise of accurate traffic state prediction is the traffic state data obtained by a large number of traffic state detectors, so the cost of a large number of detectors and the cost of maintaining the detectors are very high

Images