Predicting traffic phases from car sensor data using machine learning

E. Heyns, S. Uniyal, E. Dugundji, F. Tillema, C. Huijboom

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

This research is an explorative study to look for the potential to predict traffic density from driver behaviour using signals collected from the Controller Area Network (CAN) bus. The hypothesis is that driver behaviour is influenced by traffic density in such a way that an approximation of the traffic density can be determined from changes in the driver behaviour. Machine learning will be employed to correlate a selection of commonly available sensors on cars to the traffic density. Challenges in the processing of the data for this purpose will be outlined. The data for this study is collected from five passenger cars and nineteen trucks driving on the A28 highway in Utrecht region in the Netherlands. This study is restricted to straight roads in order to isolate the steering behaviour attributable to the traffic state influences rather than following the curve in the road. The results are encouraging that the correlation between driver behaviour and traffic density can be established. An overall accuracy of over 95% is achieved with a precision of 92%. The recall rate however is low most likely caused by over-fitting due to the unbalanced data set. The results still look promising and more training data should improve the results. This research is part of the broader project VIA NOVA which aims to investigate the use of car-sensor data for traffic and road asset management.

Original languageEnglish
Pages (from-to)92-99
Number of pages8
JournalProcedia Computer Science
Volume151
Early online date21 May 2019
DOIs
Publication statusPublished - 2019
Event10th International Conference on Ambient Systems, Networks and Technologies, ANT 2019 and The 2nd International Conference on Emerging Data and Industry 4.0, EDI40 2019, Affiliated Workshops - Leuven, Belgium
Duration: 29 Apr 20192 May 2019

Fingerprint

Learning systems
Railroad cars
Asset management
Sensors
Passenger cars
Trucks
Controllers
Processing

Bibliographical note

Part of special issue: The 10th International Conference on Ambient Systems, Networks and Technologies (ANT 2019) / The 2nd International Conference on Emerging Data and Industry 4.0 (EDI40 2019) / Affiliated Workshops. Edited by Elhadi Shakshuki

Keywords

  • CAN bus
  • Congestion
  • Driver behavior
  • Probe vehicle data
  • Supervised machine learning
  • Traffic density

Cite this

Heyns, E. ; Uniyal, S. ; Dugundji, E. ; Tillema, F. ; Huijboom, C. / Predicting traffic phases from car sensor data using machine learning. In: Procedia Computer Science. 2019 ; Vol. 151. pp. 92-99.
@article{4264340392cd4ee4aca5b5c94efaa9a4,
title = "Predicting traffic phases from car sensor data using machine learning",
abstract = "This research is an explorative study to look for the potential to predict traffic density from driver behaviour using signals collected from the Controller Area Network (CAN) bus. The hypothesis is that driver behaviour is influenced by traffic density in such a way that an approximation of the traffic density can be determined from changes in the driver behaviour. Machine learning will be employed to correlate a selection of commonly available sensors on cars to the traffic density. Challenges in the processing of the data for this purpose will be outlined. The data for this study is collected from five passenger cars and nineteen trucks driving on the A28 highway in Utrecht region in the Netherlands. This study is restricted to straight roads in order to isolate the steering behaviour attributable to the traffic state influences rather than following the curve in the road. The results are encouraging that the correlation between driver behaviour and traffic density can be established. An overall accuracy of over 95{\%} is achieved with a precision of 92{\%}. The recall rate however is low most likely caused by over-fitting due to the unbalanced data set. The results still look promising and more training data should improve the results. This research is part of the broader project VIA NOVA which aims to investigate the use of car-sensor data for traffic and road asset management.",
keywords = "CAN bus, Congestion, Driver behavior, Probe vehicle data, Supervised machine learning, Traffic density",
author = "E. Heyns and S. Uniyal and E. Dugundji and F. Tillema and C. Huijboom",
note = "Part of special issue: The 10th International Conference on Ambient Systems, Networks and Technologies (ANT 2019) / The 2nd International Conference on Emerging Data and Industry 4.0 (EDI40 2019) / Affiliated Workshops. Edited by Elhadi Shakshuki",
year = "2019",
doi = "10.1016/j.procs.2019.04.016",
language = "English",
volume = "151",
pages = "92--99",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",

}

Predicting traffic phases from car sensor data using machine learning. / Heyns, E.; Uniyal, S.; Dugundji, E.; Tillema, F.; Huijboom, C.

In: Procedia Computer Science, Vol. 151, 2019, p. 92-99.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Predicting traffic phases from car sensor data using machine learning

AU - Heyns, E.

AU - Uniyal, S.

AU - Dugundji, E.

AU - Tillema, F.

AU - Huijboom, C.

N1 - Part of special issue: The 10th International Conference on Ambient Systems, Networks and Technologies (ANT 2019) / The 2nd International Conference on Emerging Data and Industry 4.0 (EDI40 2019) / Affiliated Workshops. Edited by Elhadi Shakshuki

PY - 2019

Y1 - 2019

N2 - This research is an explorative study to look for the potential to predict traffic density from driver behaviour using signals collected from the Controller Area Network (CAN) bus. The hypothesis is that driver behaviour is influenced by traffic density in such a way that an approximation of the traffic density can be determined from changes in the driver behaviour. Machine learning will be employed to correlate a selection of commonly available sensors on cars to the traffic density. Challenges in the processing of the data for this purpose will be outlined. The data for this study is collected from five passenger cars and nineteen trucks driving on the A28 highway in Utrecht region in the Netherlands. This study is restricted to straight roads in order to isolate the steering behaviour attributable to the traffic state influences rather than following the curve in the road. The results are encouraging that the correlation between driver behaviour and traffic density can be established. An overall accuracy of over 95% is achieved with a precision of 92%. The recall rate however is low most likely caused by over-fitting due to the unbalanced data set. The results still look promising and more training data should improve the results. This research is part of the broader project VIA NOVA which aims to investigate the use of car-sensor data for traffic and road asset management.

AB - This research is an explorative study to look for the potential to predict traffic density from driver behaviour using signals collected from the Controller Area Network (CAN) bus. The hypothesis is that driver behaviour is influenced by traffic density in such a way that an approximation of the traffic density can be determined from changes in the driver behaviour. Machine learning will be employed to correlate a selection of commonly available sensors on cars to the traffic density. Challenges in the processing of the data for this purpose will be outlined. The data for this study is collected from five passenger cars and nineteen trucks driving on the A28 highway in Utrecht region in the Netherlands. This study is restricted to straight roads in order to isolate the steering behaviour attributable to the traffic state influences rather than following the curve in the road. The results are encouraging that the correlation between driver behaviour and traffic density can be established. An overall accuracy of over 95% is achieved with a precision of 92%. The recall rate however is low most likely caused by over-fitting due to the unbalanced data set. The results still look promising and more training data should improve the results. This research is part of the broader project VIA NOVA which aims to investigate the use of car-sensor data for traffic and road asset management.

KW - CAN bus

KW - Congestion

KW - Driver behavior

KW - Probe vehicle data

KW - Supervised machine learning

KW - Traffic density

UR - http://www.scopus.com/inward/record.url?scp=85071905145&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071905145&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2019.04.016

DO - 10.1016/j.procs.2019.04.016

M3 - Article

VL - 151

SP - 92

EP - 99

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -