Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Road traffic prediction method and device based on graph convolution analysis

A flow forecasting and flow technology, applied in forecasting, neural learning methods, instruments, etc., can solve problems such as seldom considering the flow connection relationship, and the flow cannot be effectively dealt with, and achieve the effect of avoiding security risks.

Pending Publication Date: 2021-03-19
NANJING ZNV SOFTWARE CO LTD +1
View PDF9 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods require the support of a large amount of historical data, and at the same time seldom consider the connection relationship between various traffic, and cannot effectively deal with unexpected events affecting traffic

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
  • Road traffic prediction method and device based on graph convolution analysis
  • Road traffic prediction method and device based on graph convolution analysis
  • Road traffic prediction method and device based on graph convolution analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0033] Such as figure 2 As shown, the road flow prediction method based on graph convolution analysis provided by the embodiment of the present application, an implementation thereof, includes the following steps:

[0034] Step 202: Determine the node for predicting the traffic service, and collect the characteristic value of the node.

[0035] Step 204: preset a loss function, use the feature value as input data to train the model, and obtain model parameters.

[0036] Step 206: Predict the flow of the node according to the model parameters and the input data.

[0037] The road flow prediction method based on graph convolution analysis provided by the embodiment of the present application can accurately predict subsequent flow data according to the connection relationship between road nodes by inputting the known road network connection structure and short-term historical flow data. Effectively predict the impact of unexpected events on traffic, so as to take effective cou...

Embodiment 2

[0056] Such as Image 6 As shown, the road flow prediction device based on graph convolution analysis provided in Embodiment 2 of the present application, an implementation manner thereof, includes a collection module 610 , a training module 620 and a prediction module 630 .

[0057] The collection module 610 is configured to determine a node for predicting traffic services, and collect characteristic values ​​of the nodes.

[0058] The training module 620 is configured to preset a loss function, use the feature value as input data to train the model, and obtain model parameters.

[0059] The prediction module 630 is configured to predict the flow of the node according to the model parameters and the input data.

[0060] The road flow prediction device based on graph convolution analysis provided by the embodiment of the present application uses model training and prediction, by inputting the known road network connection structure and short-term historical flow data, and acc...

Embodiment 3

[0072] An embodiment of the road flow prediction device based on graph convolution analysis provided in Embodiment 3 of the present application includes a memory and a processor.

[0073] Memory, used to store programs.

[0074] The processor is configured to implement the method in the first embodiment by executing the program stored in the memory.

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

The invention discloses a road traffic prediction method and device based on graph convolution analysis, and the method comprises the steps: determining a node for predicting a traffic service, and collecting the characteristic value of the node; training a model by taking the characteristic value as an input data to obtain the model; presetting a loss function, and training the model by taking the characteristic value as input data to obtain model parameters; and predicting the traffic of the node according to the model parameters and the input data. According to the method, a sub-model training prediction mode is utilized, a known road network connection structure and short-term historical traffic data are input, subsequent traffic data is accurately predicted according to the connectionrelationship between road nodes, and the influence of sudden events on traffic can be effectively predicted, so that effective response measures are taken, and the prediction accuracy is improved. Potential safety hazards are avoided, and a basis is provided for urban intelligent treatment.

Description

technical field [0001] This application relates to smart cities, and in particular to a method and device for road flow prediction based on graph convolution analysis. Background technique [0002] Urban governance and management are not only an important part of the national governance system, but also an important carrier of the global Internet governance system and an important foundation for building a community of shared future in cyberspace. In recent years, many cities in my country have launched smart city construction pilot projects, which have effectively improved the level of public services, enhanced management capabilities, and promoted urban economic development. With the implementation of the network power strategy, the national big data strategy, the "Internet +" action plan and the continuous development of the construction of "Digital China", cities have been endowed with new connotations and new requirements, which not only promote wisdom in the traditiona...

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 Applications(China)
IPC IPC(8): G06Q10/04G06Q50/26G06N3/04G06N3/08
CPCG06Q10/04G06Q50/26G06N3/08G06N3/045
Inventor 徐高峰张星裴卫斌关淑菊
Owner NANJING ZNV SOFTWARE CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products