Traffic flow prediction method, medium and equipment based on graph-Gaussian process

Abstract

The present invention disclosed a traffic flow prediction method, medium and equipment based on the Tugos process. It has statistics on traffic data and predict the mean and variance of traffic flow based on the design prediction algorithm.The method of the present invention comprehensively considers the time and space characteristics of traffic flow. Through the process of aggregating Gaussian process to characterize the space characteristics, use the deep convolution Gaussing process to perform time characteristics of multiple map aggregation Gaussian process, so as to build a complete traffic flow model, through training through trainingYou can get road traffic flow prediction models; when the model causes insufficient predictive accuracy due to external disturbances, the model correction method of the invention model can be used for post -school correction of the model.Results show that the model constructed by the present invention can accurately predict traffic flow data and can also predict errors at the same time. The correction method can make the model have the ability to correct online and improve the adaptability of the prediction algorithm.

Description

technical field [0001] The invention belongs to the field of digital intelligent transportation, and in particular relates to a traffic flow prediction method, medium and equipment based on a graph-Gaussian process. Background technique [0002] In the past few decades, the number of cars in my country has increased year after year, and it is estimated that by 2022, the number will reach more than 300 million. At the same time, the traffic demand is also increasing day by day, which makes the current road traffic load increase day by day, which brings a series of problems such as congestion and accidents. Although the traffic control department has taken measures to alleviate traffic congestion to a certain extent, such as road construction, vehicle number restriction, etc., the traffic congestion has not been improved. [0003] Traffic flow forecasting and control is the core problem of traffic efficiency. Making reasonable decisions in advance according to the forecast re...

AI Technical Summary

Problems solved by technology

However, these existing advanced traffic flow prediction models have limited use of graph structures, and the feature extraction of time series also requires relatively high data richness and quality, which limits their application in areas where data collection is sparse and scarce. No prediction uncertainty output, no model correction method, so poor adaptability in real-world scenarios

Images