Active traffic management aims to dynamically manage congestion based on existing and predicted traffic conditions. A challenge in this is that it is not usually possible to process data in real-time and use the output in control algorithms or in traveler information systems. A solution to this is to predict the traffic state based on assessments of current and past measurements. The work described in this paper develops an adaptive forecasting method to predict traffic speeds using dynamic linear models with Bayesian inference from a priori distributions. This study incorporates speeds collected from radar based sensors and validates the results with data collected from Bluetooth traffic monitoring technology. The highly adaptive model is confirmed with estimated traffic speeds during inclement weather and multiple incidents.
This document was originally published by Elsevier in Procedia Computer Science. This work is provided under a Creative Commons NonCommercial-NoDerivs 4.0 License. Details regarding the use of this work can be found at: http://creativecommons.org/licenses/by-nc-nd/4.0/. doi: 10.1016/j.procs.2014.05.435
Available at: http://works.bepress.com/brian_portugais/3/