Estimation of Link Speed Using Pattern Classification of GPS Probe Car Data

  • Seung-Heon Lee
  • Byung-Wook Lee
  • Young-Kyu Yang
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3981)


In the field of Intelligent Transport Systems, there have been many attempts to predict the speed of the vehicle by adopting the pattern of the speed using artificial intelligence and statistical methods. Traffic information has been collected mainly by fixed devices which are costly and hard to maintain. Recently, traffic data is progressively being collected by the probe cars equipped with GPS receivers. Most of probe cars are comprised of public and commercial transportation such as taxis and buses since the private drivers are reluctant to give their location information due to the privacy issues. This creates problem of insufficient number of cars available and the traditional analysis methods used for analyzing the data collected by the fixed devices are not applicable. The aim of this research is to propose and test a new method of calculating the optimal link speed for the collected information from probe cars. We propose the adoption of a fuzzy c-mean method for this purpose. In this paper the GPS speed data are automatically classified into three groups of speed patterns such as low, middle, and high speed and the link speed is predicted from the pattern clusters. In performance tests, the proposed method provides significantly better results than normal average speed data.


Traffic Information Membership Grade Intelligent Transport System Transportation Research Part Link Travel Time 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hoppner, F.: A Contribution to Convergence Theory of Fuzzy c-Mean and Derivatives. IEEE Transactions of Fuzzy System, 11(5) (2003)Google Scholar
  2. 2.
    You, J., Kim, T.J.: Development and evaluation of a hybrid travel time forecasting model. Transportation Research Part C: Emerging Technologies 8(1-6), 231–256 (2000)CrossRefGoogle Scholar
  3. 3.
    Coifman, B.: Estimating travel times and vehicle trajectories on freeways using dual loop detectors. Transportation Research: Part A 36(4), 351–364 (2002)Google Scholar
  4. 4.
    Dharia, A., Adeli, H.: Neural network model for rapid forecasting of freeway link travel time. Engineering Applications of Artificial Intelligence 16(7-8), 607–613 (2003)CrossRefGoogle Scholar
  5. 5.
    Chan, K.S., Lam, W.H.K.: Optimal speed detector density for the network with travel time information. Transportation Research Part A: Policy and Practice 36(3), 203–223 (2002)CrossRefGoogle Scholar
  6. 6.
    Wang, Y., Papageorgiou, M.: Real-time freeway traffic state estimation based on extended Kalman filter: a general approach. Transportation Research Part B: Methodological 39(2), 141–167 (2005)CrossRefGoogle Scholar
  7. 7.
    Hellinga, B., Fu, L.: Reducing bias in probe-based arterial link travel time estimates. Transportation Research Part C: Emerging Technologies 10(4), 257–273 (2002)CrossRefGoogle Scholar
  8. 8.
    Goktepe, A.B., Altun, S., Sezer, A.: Soil clustering by fuzzy c-means algorithm. Advances in Engineering Software 36(10), 691–698 (2005)CrossRefGoogle Scholar
  9. 9.
    Mingoti, S.A., Lima, J.: Comparing SOM neural network with Fuzzy c-means, K-means and traditional hierarchical clustering algorithms. European Journal of Operational Research. In Press, Corrected Proof (2005)Google Scholar
  10. 10.
    Choi, K.J., Chung, Y.S.: A Data Fusion Algorithm for Estimating Link Travel Time. Intelligent Transportation Systems Journal 7(3-4), 235–260 (2002)MATHGoogle Scholar
  11. 11.
    Boyce, D., Rouphail, N., Kirson, A.: Estimation and measurement of link travel times in ADVANCE project. In: Proceedings of Vehicle Navigation and Information Systems Conference, pp. 62–66. IEEE, Los Alamitos (1993)CrossRefGoogle Scholar
  12. 12.
    Sisiopiku, V., Rouphail, N.: Exploratory Analysis of the Correlations Between Arterial link Travel Times and Detector Data from Simulation and Field Studies. In: Advance Working Paper No. 25 (October 1993)Google Scholar
  13. 13.
    Sisiopiku, V., Palacharla, P., Nelson, P.: Fuzzy Reasoning Model for Converting Loop Detector Data into Travel Times. In: Advance Working Paper Series No. 38 (June 1994)Google Scholar
  14. 14.
    Nelson, P., Palacharla, P.: A neural network model for data fusion in advance. In: Proceedings of Pacific Rim Conference, Seattle, Wash, pp. 237–243 (1993)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Seung-Heon Lee
    • 1
  • Byung-Wook Lee
    • 1
  • Young-Kyu Yang
    • 1
  1. 1.College of SoftwareKyungwon UniversityGyeonggi-doSouth Korea

Personalised recommendations