Travel Time Estimation Algorithms Research Articles

Using the robust statistics for travel time estimation on highways

Highway operators around the world are using automated vehicle identification (AVI)-based techniques as a technological input for travel time estimation on highways. Various AVI technologies provide various travel time measurement samples: some of them are able to identify only personal cars (e.g. tolling tags), while others provide mixed samples of all vehicle classes (e.g. license plate matching). As the adequate information on travel times should concern the personal cars, the influence of heavy vehicles (HVs) should be eliminated from the samples, which is not feasible with the use of existing travel time estimation algorithms. It was observed that also during congestion travel times of personal cars and HVs remain dispersed. The motivation for the present study was to introduce an algorithm that would be able to exclude the influence of slower HVs in travel time estimation for technologies, providing mixed samples of travel time measurements. This was achieved by the use of robust statistics. The results of the study could be used by all highway agencies and operators who are encountering problems with unreasonably extended estimations of travel times because of the presence of slow HVs in the traffic flow.

Estimating Required Probe Vehicle Re-identification Requirements for Characterizing Link Travel Times

In the 1970s a framework was developed by Oppenlander to determine the sample size required for travel time estimation studies. This framework is still recommended today. This paper develops a new framework to improve upon the ideas set forth by Oppenlander. This new framework is based upon the Kullback-Leibler divergence. It allows for travel time studies to be evaluated in a more comprehensive way. Travel time estimation methods can now be evaluated on their ability to estimate travel time distributions instead of only the mean travel time. Also, this framework can be used on any travel time distribution whereas the Oppenlander framework was only properly suited for Gaussian distributions. The Kullback-Leibler divergence also allows for comparing both ID matching (i.e., small sample sizes with no erroneous travel times) and signature matching (i.e., large sample sizes mixed with some erroneous travel times) travel time estimation algorithms to be evaluated, while the Oppenlander framework was best suited for the ID matching algorithms. In this paper the Kullback-Leibler comparison framework for travel time studies is developed. The framework is then used to provide a comparison of an example ID matching and an example signature matching algorithm to demonstrate how both can be evaluated in a single framework. Finally, conclusions are made about the usefulness of the Kullback-Leibler comparison framework.

Sensing and Signal Processing for Vehicle Reidentification and Travel Time Estimation

Link travel times are crucial for advanced traveler information systems and traffic management applications. However, current systems for estimating them still have shortcomings that need to be addressed. In this paper, we propose a novel framework for vehicle reidentification via signature matching using signal processing techniques and a travel time estimation algorithm that is robust to potential (and often inevitable) vehicle misidentifications. Individual vehicles are matched between well-separated stations in a road transportation network using signatures captured by embedded roadway sensors. Statistical and multirate signal processing methods are used to develop data-postprocessing algorithms that are critical to the subsequent signature-matching problem, which is formulated using optimal techniques from communication theory. A probabilistic modeling of the generated matching assignments and an unsupervised data-clustering technique are then used to devise a travel time estimation algorithm. The proposed method is tested under a real traffic scenario, and accurate link travel time measures are reported.

Highway travel time accurate measurement and short-term prediction using multiple data sources

The development of new traffic monitoring systems and the increasing interest of road operators and researchers in obtaining reliable travel time measurements, motivated by society's demands, have led to the development of multiple travel time data sources and estimation algorithms. This situation provides a perfect context for the implementation of data fusion methodologies to obtain the maximum accuracy from the combination of the available data. This article presents a new and simple approach for the short-term prediction of highway travel times, which represent an accurate estimation of the expected travel time for a driver commencing on a particular route. The algorithm is based on the fusion of different types of data that come from different sources (inductive loop detectors and toll tickets) and from different calculation algorithms. Although the data fusion algorithm presented herein is applied to these particular sources of data, it could easily be generalised to other equivalent types of data. The objective of the proposed data fusion process is to obtain a fused value more reliable and accurate than any of the individual estimations. The methodology overcomes some of the limitations of travel time estimation algorithms based on unique data sources, as the limited spatial coverage of the algorithms based on spot measurement or the information delay of direct travel time itinerary measurements when disseminating the information to the drivers in real time. The results obtained in the application of the methodology on the AP-7 highway, near Barcelona in Spain, are found to be reasonable and accurate. In short, the travel time data fusion algorithm presented in this article tries to be as simple as possible and yet still improve the existing naïve approaches.

Estimation of the automatic vehicle identification based spatial travel time information collected in Stockholm

To support the implementation of real-time traffic information systems in the Stockholm city area using automatic vehicle identification (AVI) data, a preliminary travel time analysis tool has been developed. The program can manage and analyse travel time measurements in a distributed database server where both online and historical traffic information are saved. Meanwhile, several existing travel time estimation algorithms are implemented in the travel time analysis program, and are evaluated using four months of AVI data collected in the urban streets and arterials of and near the Stockholm downtown area. The advantages and disadvantages of those algorithms are also analysed using the highly noisy travel time measurements collected under the urban context. In addition, the authors have also evaluated a common statistical median filtering approach and suggested some modifications for AVI data estimation. In general, all these algorithms have the potential to be applied for real daily travel time estimation and the statistical median filter with modifications has been suggested for historical travel time estimation in real application. Finally, the authors point out an essential problem in travel time estimation and suggest a direction that may have the potential to improve the online traffic information quality.

USING AUTOMATIC VEHICLE IDENIFICATION DATA FOR TRAVEL TIME ESTIMATION IN HONG KONG

This paper presents a Real-time Traveler Information System (RTIS) for Hong Kong, in which a novel solution algorithm is proposed for estimation of both the current and instantaneous travel times using automatic vehicle identification (AVI) data. The proposed algorithm also deduces the travel times on the other road links without real-time AVI data. By integration of the filtered real-time and off-line database together, the RTIS can provide area-wide traffic information in the whole network of Hong Kong. The travel time estimates are updated once every five minutes in the RTIS. Observation surveys are conducted during different time periods at a selected path in Hong Kong urban area to validate the RTIS travel time estimates. With the use of the same set of observation data, comparison is also given to the estimation results of the other three existing AVI travel time estimation algorithms: TransGuide, TranStar and Transmit algorithms in US. It was found that the proposed RTIS algorithm could generate the current travel time estimates for the three study periods with the minimum absolute errors and absolute percentage errors. The estimations of instantaneous travel times on the road sections with and without real-time AVI data have also been illustrated. The validation results show that the performance of RTIS is satisfactory and acceptable.