Prediction device and prediction method for short time traffic conditions

Abstract

The invention discloses a prediction device for short time traffic conditions, comprising: a traffic conditions analyzing unit for reading historical traffic conditions data, and filtering the historical traffic conditions data according to a predetermined time interval to obtain the filtered historical traffic conditions data; a time sequence division unit for dividing the filtered historical traffic conditions data into a plurality of traffic conditions data according to the time sequence; and a traffic conditions prediction unit for generating a prediction result of future traffic conditions according to the plurality of traffic conditions data and a fuzzy average predetermined prediction model. The invention further discloses a prediction method for short time traffic conditions. With the method and the device provided by the invention, the plurality of traffic conditions data are divided from the historical traffic conditions data, characters of traffic conditions of each traffic conditions data are accurately represented, and influence of non-steady variation of the traffic conditions on the prediction result is reduced by a fuzzy average prediction method, so that large-scale application of the method and the device is implemented.

Description

technical field [0001] The present invention relates to the technical field of traffic management equipment, in particular to a traffic state prediction device and a prediction method, in particular to a short-term segmented traffic state prediction device and prediction method. Background technique [0002] In the intelligent urban traffic management technology, technical means such as traffic flow guidance, traffic signal control, and event detection are often used. The real-time prediction data of various parameters that characterize the traffic state are the necessary basis for realizing the above technical means. Therefore, the real-time prediction of various parameters of the traffic state is the premise and key to realize informatization, modernization, intelligence, and integration of traffic management. The accuracy of the prediction results of traffic state parameters is directly related to the real-time performance of intelligent induction and control in traffic ma...

AI Technical Summary

Problems solved by technology

[0006] The technical problem to be solved by the present invention is to overcome the shortcomings of the existing prediction method that the collection parameters are relatively simple and cannot accurately represent the characteristics of traffic state changes, and to provide a traffic state prediction device, which can accurately Characterize the traffic state characteristics of each time period to select different traffic signal control strategies for different traffic state data, thereby improving the prediction accuracy of future traffic states

Images