Large scale estimation of arterial traffic and structural analysis of traffic patterns using probe vehicles

Aude Hofleitner; Ryan Herring; Alexandre Bayen; Yufei Han; Fabien Moutarde; Arnaud de La Fortelle

Large scale estimation of arterial traffic and structural analysis of traffic patterns using probe vehicles

Aude Hofleitner ¹ Ryan Herring ² Alexandre Bayen ³ Yufei Han ⁴ Fabien Moutarde ⁴ Arnaud de La Fortelle ^{4, 5}

1 EECS - Electrical Engineering and Computer Science [Berkeley]

2 California Center for Innovative Transportation,

3 CEE - Department of Civil and Environmental Engineering [Berkeley]

4 CAOR - Centre de Robotique

5 IMARA - Informatique, Mathématiques et Automatique pour la Route Automatisée

Inria Paris-Rocquencourt

Abstract : Estimating and analyzing traffi c conditions on large arterial networks is an inherently diffi cult task. The fi rst goal of this article is to demonstrate how arterial tra c conditions can be estimated using sparsely sampled GPS probe vehicle data provided by a small percentage of vehicles. Traffi c signals, stop signs, and other flow inhibitors make estimating arterial traffi c conditions significantly more diffi cult than estimating highway traffi c conditions. To address these challenges, we propose a statistical modeling framework that leverages a large historical database and relies on the fact that tra ffic conditions tend to follow distinct patterns over the course of a week. This model is operational in North California, as part of the Mobile Millennium tra ffic estimation platform. The second goal of the article is to provide a global network-level analysis of tra ffic patterns using matrix factorization and clustering methods. These techniques allow us to characterize spatial tra ffic patterns in the network and to analyze traffi c dynamics at a network scale. We identify tra ffic patterns that indicate intrinsic spatio-temporal characteristics over the entire network and give insight into the traffi c dynamics of an entire city. By integrating our estimation technique with our analysis method, we achieve a general framework for extracting, processing and interpreting traffi c information using GPS probe vehicle data.

Type de document :

Communication dans un congrès

Domaine :

Statistiques [stat] / Machine Learning [stat.ML]

Liste complète des métadonnées

Littérature citée [4 références] Voir Masquer Télécharger

https://hal-mines-paristech.archives-ouvertes.fr/hal-00741497
Contributeur : Fabien Moutarde <>
Soumis le : vendredi 12 octobre 2012 - 23:45:19
Dernière modification le : lundi 12 novembre 2018 - 11:02:33
Document(s) archivé(s) le : samedi 17 décembre 2016 - 00:38:24

main_trb2012_Gaussian_NMF.pdf

Fichiers produits par l'(les) auteur(s)

Large scale estimation of arterial traffic and structural analysis of traffic patterns using probe vehicles

Fichier

Identifiants

Collections

Citation

Exporter

Métriques

535

424

Contact

Large scale estimation of arterial traffic and structural analysis of traffic patterns using probe vehicles

Fichier

Identifiants

Collections

Citation

Exporter

Partager

Métriques

535

424

Contact