Travel Time Forecasting and Dynamic Origin-Destination Estimation for Freeways Based on Bluetooth Traffic Monitoring

Abstract

From the point of view of the information supplied by an ATIS to the motorists entering a freeway of one of the most relevant is the Forecasted Travel Time, that is the expected travel time that they will experience when traverse a freeway segment. From the point of view of ATMS the dynamic estimates of time dependencies in OD matrices is a major input to dynamic traffic models used for estimating the current traffic state and forecasting its short term evolution. Travel Time Forecasting and Dynamic OD Estimation are two of the key components of ATIS/ATMS and the quality of the results that they could provide depend not only on the quality of the models but also on the accuracy and reliability of the measurements of traffic variables supplied by the detection technology.The quality and reliability of the measurements produced by traditional technologies, as inductive loop detectors, is not usually the required by real-time applications, therefore one wonders what could be expected from the new ICT technologies as for example Automatic Vehicle Location, License Plate Recognition, detection of mobile devices and so on. The main objectives of this paper are: to explore the quality of the data produced by the Bluetooth detection of mobile devices equipping vehicles for Travel Time Forecasting and its use to estimate time dependent OD matrices. Ad hoc procedures based on Kalman Filtering have been designed and implemented successfully and the numerical results of the computational experiments are presented and discussed.

A Kalman Filter Approach for Exploiting Bluetooth Traffic Data When Estimating Time-Dependent OD Matrices

“…This clean data is the main input to a new Kalman Filter approach for estimating timedependent OD matrices. The proposed approach, which exploits the explicit travel time measurements from Bluetooth detectors, is based on a reformulation of the Kalman Filter approach for freeways explored in Barceló et al (2010a). It also extends the approach to urban networks where alternative paths are available and route choice is relevant.…”

“…Then, they are adjusted from the available link counts provided by an existing layout of traffic counting stations and other additional information whenever it is available. Adjustments can be considered as indirect estimation methods, based either on discrete time optimization approaches (Codina & Barceló (2004); Lundgren & Peterson (2008)) or on adaptations of Kalman Filtering approaches (Ashok & Ben Akiva, 2000;Antoniou, BenAkiva & Koutsopoulos, 2007;Barceló et al 2010a We placed flow counting detectors and ICT sensors in a cordon and at each possible point for flow entry (centroids of the study area). ICT sensors were located at intersections in urban networks and covered access and links to/from the intersection.…”

“…The final two columns correspond to the calculated speed and travel time. Data was collected over the two months of May and June 2009, and it was used to create a Historic Database of past measurements and traffic patterns which, together with real-time detection, provided the input for both the filtering and the forecasting algorithm (Barceló et al 2010a). …”

New Tools for Studying Visitor Behaviours in Museums: A Case Study at the Louvre

“…The recent wide spread of mobile devices implies that many people have their Bluetooth switched on passively, thus providing an important source of useful data. A variety of projects have exploited Bluetooth data for measuring the social network relationships between people (Eagle & Pentland, 2005, Paulos & Goodman, 2004, Nicolai, Yoneki, Behrens, & Kenn, 2006, mobility of vehicles (Yalowitz & Bronnenkant, 2009, Barceló, Montero, Marqués, & Carmona, 2010 and mobility of pedestrians and their relationships (O'Neill et al, 2006, Kostakos et al, 2010. However these investigations have not considered a specific analysis of pedestrians and their use of space.…”

Hidden Markov Flow Network Model: A Generative Model for Dynamic Flow on a Network

“…Traditionally, time dependent OD matrices were estimated by methods such as data assimilation using observed traffic counts on roads [2], [3]. Recently, however, because more and more GPS and communication devices are being equipped on vehicles, we can obtain a large, high-resolution dynamic OD matrix which reflects spatio-temporal mobility patterns.…”