Computing travel times from filtered traffic states

Pierre Emmanuel Mazaré, Olli Pekka Tossavainen, Daniel B. Work

Research output: Contribution to journalArticlepeer-review


This article experimentally assesses the influence of sensor data rates on travel time estimates computed from filtered traffic speed estimates. Using velocity data obtained from GPS smartphones and inductive loop detector data collected during the Mobile Century experiment near Berkeley, CA, and an evolution equation for average velocity along the roadway, an estimate of the traffic state is obtained via ensemble Kalman filtering. A large-scale batch of computations is run to produce estimates of traffic velocity with varying degrees of input data, and instantaneous and a posteriori dynamic travel times are compared to travel times recorded using license plate re-identification. We illustrate that dynamic travel time estimates can be computed with less than 10% error regardless of the data source, and that existing inductive loop detector data can significantly improve the accuracy of travel time estimates when GPS data is sparse.

Original languageEnglish (US)
Pages (from-to)557-578
Number of pages22
JournalDiscrete and Continuous Dynamical Systems - Series S
Issue number3
StatePublished - Jun 2014


  • Experimental data
  • Scalar conservation laws
  • Traffic estimation
  • Traffic flow
  • Vehicular traffic

ASJC Scopus subject areas

  • Analysis
  • Discrete Mathematics and Combinatorics
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Computing travel times from filtered traffic states'. Together they form a unique fingerprint.

Cite this