IMPROVED KALMAN FILTERING APPROACH FOR ESTIMATING ORIGIN-DESTINATION MATRICES FOR FREEWAY CORRIDORS

The estimation of origin-destination (OD) matrices for freeway corridors by using inner-link induction loop data is examined. A trip generation model is used, and various parameter optimization and statistics-based methods are examined to estimate the split parameters in the model. A Kalman-based method that uses the model-predicted link-flow variances and covariances while processing the measurements is described. A simple but effective solution to the problem of initializing the Kalman filter and imposing the natural constraints to the estimates is presented. The resulting method is tested on both simulated and observed data and is compared with other methods such as least squares and constrained optimization, showing that the Kalman-based method leads to the best results.

Language

  • English

Media Info

  • Features: Figures; References;
  • Pagination: p. 54-64
  • Monograph Title: Travel demand modeling and network assignment models
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 00670484
  • Record Type: Publication
  • ISBN: 0309055245
  • Files: TRIS, TRB
  • Created Date: Dec 23 1994 12:00AM