Heteroscedastic Gaussian processes for uncertainty modeling in large-scale crowdsourced traffic data |
| |
| Institution: | 1. Technical University of Denmark (DTU), Bygning 116B, 2800 Kgs. Lyngby, Denmark;2. Massachusetts Institute of Technology (MIT), 77 Mass. Ave., 02139 Cambridge, MA, USA;1. Engineering Research Center of Digitized Textile & Apparel Technology, Ministry of Education, College of Information Sciences and Technology, Donghua University, Shanghai 201620, PR China;2. Department of Chemical and Materials Engineering, University of Alberta, Canada;1. NEXTRANS Center, Purdue University, West Lafayette, IN 47906, USA;2. Goergen Institute for Data Science, University of Rochester, Rochester, NY 14627, USA;3. Department of Civil and Environmental Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, USA;4. Department of Civil Engineering and Engineering Mechanics, The University of Arizona, AZ 85721, USA;5. Department of Electrical and Computer Engineering, Ohio State University, OH 43210, USA;6. Department of Civil, Structural and Environmental Engineering, University at Buffalo, the State University of New York, Buffalo, NY 14260, USA;1. OPTIMA Unit, TECNALIA, P. Tecnologico Bizkaia, Ed. 700, 48160 Derio, Spain;2. Dept. of Communications Engineering, University of the Basque Country UPV/EHU, Alameda Urquijo S/N, 48013 Bilbao, Spain;3. Basque Center for Applied Mathematics (BCAM), 48009 Bilbao, Spain;1. Department of Civil & Environmental Engineering, University of Maryland, College Park, MD 20740, United States;2. Zachry Department of Civil Engineering, Texas A&M University, College Station, TX 77843-3136, United States;1. Department of Civil and Environmental Engineering, University of Tennessee, Knoxville, 851 Neyland Drive, Knoxville, TN 37996, USA;2. Department of Geography, University of Tennessee, Knoxville, 1000 Phillip Fulmer Way, Knoxville, TN 37916, USA;3. Center for Transportation Analysis, Oak Ridge National Laboratory, 2360 Cherahala Blvd, Knoxville, TN 37932, USA;4. Tennessee Department of Transportation, James K. Polk Building, 505 Deaderick Street, Suite 300, Nashville, TN 37243, USA;1. Department of Civil, Environmental and Construction Engineering, University of Central Florida, Orlando, FL 32816, USA;2. Department of Transportation and Logistics Engineering, Hanyang University, Ansan 15588, Republic of Korea |
| |
| Abstract: | Accurately modeling traffic speeds is a fundamental part of efficient intelligent transportation systems. Nowadays, with the widespread deployment of GPS-enabled devices, it has become possible to crowdsource the collection of speed information to road users (e.g. through mobile applications or dedicated in-vehicle devices). Despite its rather wide spatial coverage, crowdsourced speed data also brings very important challenges, such as the highly variable measurement noise in the data due to a variety of driving behaviors and sample sizes. When not properly accounted for, this noise can severely compromise any application that relies on accurate traffic data. In this article, we propose the use of heteroscedastic Gaussian processes (HGP) to model the time-varying uncertainty in large-scale crowdsourced traffic data. Furthermore, we develop a HGP conditioned on sample size and traffic regime (SSRC-HGP), which makes use of sample size information (probe vehicles per minute) as well as previous observed speeds, in order to more accurately model the uncertainty in observed speeds. Using 6 months of crowdsourced traffic data from Copenhagen, we empirically show that the proposed heteroscedastic models produce significantly better predictive distributions when compared to current state-of-the-art methods for both speed imputation and short-term forecasting tasks. |
| |
| Keywords: | Gaussian processes Heteroscedastic models Traffic data Crowdsourcing Uncertainty modeling Forecasting Imputation Floating car data |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
