Predicting arrival times of buses using real-time GPS measurements

Mathieu Sinn, Ji Won Yoon, Francesco Calabrese, Eric Bouillet

Research output: Chapter in Book/Report/Conference proceedingConference contribution

25 Citations (Scopus)

Abstract

Predicting arrival times of buses is a key challenge in the context of building intelligent public transportation systems. In this paper, we describe an efficient non-parametric algorithm which provides highly accurate predictions based on real-time GPS measurements. The key idea is to use a Kernel Regression model to represent the dependencies between position updates and the arrival times at bus stops. The performance of the proposed algorithm is evaluated on real data from the public bus transportation system in Dublin, Ireland. For a time horizon of 50 minutes, the prediction error of the algorithm is less than 10 percent on average. It clearly outperforms parametric methods which use a Linear Regression model, predictions based on the K-Nearest Neighbor algorithm, and a system which computes predictions of arrival times based on the current delay of buses. A study investigating the selection of interpolation points to reduce the size of the training set concludes the paper.

Original languageEnglish
Title of host publicationIEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC
Pages1227-1232
Number of pages6
DOIs
Publication statusPublished - 2012 Dec 21
Externally publishedYes
Event2012 15th International IEEE Conference on Intelligent Transportation Systems, ITSC 2012 - Anchorage, AK, United States
Duration: 2012 Sep 162012 Sep 19

Other

Other2012 15th International IEEE Conference on Intelligent Transportation Systems, ITSC 2012
CountryUnited States
CityAnchorage, AK
Period12/9/1612/9/19

Fingerprint

Global positioning system
Bus transportation
Intelligent buildings
Linear regression
Interpolation

ASJC Scopus subject areas

  • Automotive Engineering
  • Mechanical Engineering
  • Computer Science Applications

Cite this

Sinn, M., Yoon, J. W., Calabrese, F., & Bouillet, E. (2012). Predicting arrival times of buses using real-time GPS measurements. In IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC (pp. 1227-1232). [6338767] https://doi.org/10.1109/ITSC.2012.6338767

Predicting arrival times of buses using real-time GPS measurements. / Sinn, Mathieu; Yoon, Ji Won; Calabrese, Francesco; Bouillet, Eric.

IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC. 2012. p. 1227-1232 6338767.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Sinn, M, Yoon, JW, Calabrese, F & Bouillet, E 2012, Predicting arrival times of buses using real-time GPS measurements. in IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC., 6338767, pp. 1227-1232, 2012 15th International IEEE Conference on Intelligent Transportation Systems, ITSC 2012, Anchorage, AK, United States, 12/9/16. https://doi.org/10.1109/ITSC.2012.6338767
Sinn M, Yoon JW, Calabrese F, Bouillet E. Predicting arrival times of buses using real-time GPS measurements. In IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC. 2012. p. 1227-1232. 6338767 https://doi.org/10.1109/ITSC.2012.6338767
Sinn, Mathieu ; Yoon, Ji Won ; Calabrese, Francesco ; Bouillet, Eric. / Predicting arrival times of buses using real-time GPS measurements. IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC. 2012. pp. 1227-1232
@inproceedings{821c63e817344b40b9904c128333b3b7,
title = "Predicting arrival times of buses using real-time GPS measurements",
abstract = "Predicting arrival times of buses is a key challenge in the context of building intelligent public transportation systems. In this paper, we describe an efficient non-parametric algorithm which provides highly accurate predictions based on real-time GPS measurements. The key idea is to use a Kernel Regression model to represent the dependencies between position updates and the arrival times at bus stops. The performance of the proposed algorithm is evaluated on real data from the public bus transportation system in Dublin, Ireland. For a time horizon of 50 minutes, the prediction error of the algorithm is less than 10 percent on average. It clearly outperforms parametric methods which use a Linear Regression model, predictions based on the K-Nearest Neighbor algorithm, and a system which computes predictions of arrival times based on the current delay of buses. A study investigating the selection of interpolation points to reduce the size of the training set concludes the paper.",
author = "Mathieu Sinn and Yoon, {Ji Won} and Francesco Calabrese and Eric Bouillet",
year = "2012",
month = "12",
day = "21",
doi = "10.1109/ITSC.2012.6338767",
language = "English",
isbn = "9781467330640",
pages = "1227--1232",
booktitle = "IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC",

}

TY - GEN

T1 - Predicting arrival times of buses using real-time GPS measurements

AU - Sinn, Mathieu

AU - Yoon, Ji Won

AU - Calabrese, Francesco

AU - Bouillet, Eric

PY - 2012/12/21

Y1 - 2012/12/21

N2 - Predicting arrival times of buses is a key challenge in the context of building intelligent public transportation systems. In this paper, we describe an efficient non-parametric algorithm which provides highly accurate predictions based on real-time GPS measurements. The key idea is to use a Kernel Regression model to represent the dependencies between position updates and the arrival times at bus stops. The performance of the proposed algorithm is evaluated on real data from the public bus transportation system in Dublin, Ireland. For a time horizon of 50 minutes, the prediction error of the algorithm is less than 10 percent on average. It clearly outperforms parametric methods which use a Linear Regression model, predictions based on the K-Nearest Neighbor algorithm, and a system which computes predictions of arrival times based on the current delay of buses. A study investigating the selection of interpolation points to reduce the size of the training set concludes the paper.

AB - Predicting arrival times of buses is a key challenge in the context of building intelligent public transportation systems. In this paper, we describe an efficient non-parametric algorithm which provides highly accurate predictions based on real-time GPS measurements. The key idea is to use a Kernel Regression model to represent the dependencies between position updates and the arrival times at bus stops. The performance of the proposed algorithm is evaluated on real data from the public bus transportation system in Dublin, Ireland. For a time horizon of 50 minutes, the prediction error of the algorithm is less than 10 percent on average. It clearly outperforms parametric methods which use a Linear Regression model, predictions based on the K-Nearest Neighbor algorithm, and a system which computes predictions of arrival times based on the current delay of buses. A study investigating the selection of interpolation points to reduce the size of the training set concludes the paper.

UR - http://www.scopus.com/inward/record.url?scp=84871233803&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84871233803&partnerID=8YFLogxK

U2 - 10.1109/ITSC.2012.6338767

DO - 10.1109/ITSC.2012.6338767

M3 - Conference contribution

AN - SCOPUS:84871233803

SN - 9781467330640

SP - 1227

EP - 1232

BT - IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC

ER -