01137494

Component

This study develops a dynamic short-term corridor travel time prediction model using the traffic flow data collected by single loop detectors. This method involves a multi-step-ahead prediction of traffic condition with a seasonal autoregressive integrated moving average model. In order to adjust the prediction error based on traffic flow data that becomes available in real time, an embedded adaptive Kalman filter is designed. Then, an on-line corridor travel time prediction model is developed. Test results show that this method is able to capture the traffic dynamics and provide more accurate travel time prediction. The proposed method is promising in real-time corridor travel time prediction for various traffic applications under the context of Intelligent Transportation Systems.

01120148

09-3022

English

16p

2009

Transportation Research Board 88th Annual Meeting

DVD

Figures (6) ; Tables (4)

Estimating; Kalman filtering; Loop detectors; Traffic flow; Traffic forecasting; Traffic models; Traffic volume; Travel time

Data and Information Technology; Highways; I71: Traffic Theory

Transportation Research Board Annual Meeting 2009 Paper #09-3022

BTRIS, TRIS, TRB

Jan 30 2009 7:24PM