Unlock instant, AI-driven research and patent intelligence for your innovation.

A road network traffic data restoration method based on sae-gan-sad

A SAE-GAN-SAD, traffic data technology, applied in the field of intelligent transportation, can solve the problems of not being able to fully mine the characteristics of road traffic data data, and the accuracy of data restoration is not high, and achieve the effect of improving accuracy

Active Publication Date: 2020-12-01
ZHEJIANG UNIV OF TECH
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, most repair algorithms cannot fully mine the potential data characteristics of road traffic data, so the accuracy of data repair is not high

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A road network traffic data restoration method based on sae-gan-sad
  • A road network traffic data restoration method based on sae-gan-sad
  • A road network traffic data restoration method based on sae-gan-sad

Examples

Experimental program
Comparison scheme
Effect test

example

[0064] Example: the data in the actual experiment, the implementation plan is as follows:

[0065] (1) Select experimental data

[0066] The source of the experimental data set is the California Transportation Performance Measurement System (PeMS). The experiment selects the traffic flow data of 22 road detectors. The data sampling period is 5 minutes. The data selection time range is from May 1, 2014 to June 2014. 30 days.

[0067] The input of the model is the daily traffic flow data of 22 roads, and the missing data is simulated according to a certain missing ratio.

[0068] (2) Parameter determination

[0069] The stack autoencoder is composed of three autoencoder stacks, and the number of hidden layer units is 2048, 1024, and 512 respectively; the generator and the discriminator have the same model structure except for the output layer, and the same model structure is composed of It consists of 3 layers of neurons, the number of neurons in the hidden layer is 256, 128,...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A road network traffic data restoration method based on SAE-GAN-SAD, comprising the following steps: 1) obtaining road network traffic data, constructing a stack autoencoder and extracting features from the road traffic data; 2) determining a generator and a discriminator The structure of the generative adversarial network model is jointly constructed, and the extracted spatiotemporal features of the road traffic state are used as the input of the generator, and the loss functions of the generator and the discriminator are respectively defined, so that the generator and the discriminator can conduct confrontation training at the same time, and realize the loss function based on the missing Data spatio-temporal features generate complete data spatio-temporal features; 3) Obtain spatio-temporal features of traffic state data generated after confrontation training of generative confrontation network, use stack self-decoder to decode repaired traffic state data, and realize road traffic state data restoration. The present invention uses the SAE-GAN-SAD model to repair missing data in real time based on known traffic data, which can effectively improve the accuracy of traffic state data repair.

Description

technical field [0001] The invention relates to a road network traffic data restoration method based on SAE-GAN-SAD, and the invention belongs to the field of intelligent traffic. Background technique [0002] The integrity of road traffic flow data has a direct impact on road traffic flow prediction and real-time road regulation in intelligent transportation systems. In the real road traffic system, the problem of missing traffic flow data due to sensor failure and various irresistible factors is common. Therefore, road traffic flow data restoration is of great significance to the development of intelligent transportation systems. [0003] The current road traffic data repair methods mainly use time information or spatial relationship to repair missing data. Common algorithms for data repair using time correlation include historical average method, moving average method, exponential smoothing method, etc.; using spatial correlation Common algorithms for data restoration a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G08G1/01G06K9/62G06N3/04
CPCG08G1/0104G08G1/0125G06N3/045G06F18/214
Inventor 徐东伟魏臣臣林臻谦戴宏伟彭鹏周磊
Owner ZHEJIANG UNIV OF TECH