City expressway short-time traffic information predicting system and method

Abstract

The invention relates to a city expressway short-time traffic information predicting system and a method, which belong to the field of traffic information prediction. The system consists of a traffic information fusion sub system, a traffic information mathematical model predicting sub system and a traffic information communication and delivery sub system, wherein the traffic information fusion sub system uses the historical statistical traffic information and the real-time detection traffic information, carries out pretreatment and data fusion, and finally stores results into a collection database to be used by the predicting sub system. The traffic information mathematical model predicting sub system is based on the theory of the phase space reconstruction, adopts a chaotic time sequence self-adapting predicting method based on the time sequence reconstruction, and carries out analysis and prediction on the short-time traffic information within 15 minutes on the city expressway. The predicted traffic information is delivered in real time through the traffic information communication and delivery sub system. The invention provides the reference for the traffic management department to optimize the traffic management schemes and prevent the traffic accidents, the traffic resources are optimized, and the social benefits and the economic benefits are improved.

Description

technical field [0001] The invention relates to the field of traffic information forecasting, and in particular provides a system and method for predicting urban expressway short-term traffic information based on a chaotic time series adaptive forecasting method based on time series reconstruction. Background technique [0002] Expressway short-term traffic information prediction is the basis and key of urban expressway traffic control and guidance and other traffic management methods. Accurate and real-time short-term traffic information prediction is the premise of realizing traffic control and management and traffic guidance, and is the key to make intelligent transportation system become "active action". Traffic information service and traffic management based on predictive information belong to the frontier field of intelligent transportation system research in the world. Some developed countries and regions have carried out related research work, but good and mature ap...

AI Technical Summary

Problems solved by technology

But, there are two problems in the described traffic information prediction method: (1) the reliability problem of traffic data source

The reliability of traffic flow data plays a decisive role in the prediction of traffic information. However, the traffic information obtained often contains errors and missing data, so the use of traffic information data that is not identified and repaired reduces the reliability of model prediction.

(2) Defects of the traffic information prediction model

The neural network model has the advantages of self-adaptation and self-learning, but online update is a difficult problem, and there are specific implementation problems; the non-parametric regression model uses the mean value of similar samples in historical data as the predicted value based on sample similarity, but The implementation and reliability of forecasts are heavily dependent on complete historical data, and are not suitable for combination with other models as a sub-model

Therefore, there will be a large deviation in the actual traffic information prediction.

Images