TRAFFIC CONGESTION DETECTION FROM SUPERVISED LEARNING

Advanced Traveler Information System (ATIS) has attracted world-wide interest as a promising program to improve efficiency of road transportation systems. An optimal route guidance is an important feature of ATIS. Information regarding the level of congestion in a given roadway system is vital for deriving an optimal route guidance strategy. Characterization of level of congestion in a given roadway system requires integration of localized sensor information that can be accomplished by mapping the set of localized information into a higher dimensional space. The diagnosis of integrated information requires partitioning of the higher dimensional space into multiple regions. This paper describes a supervised learning approach, based on potential functions, to partition the higher dimensional space into regions that represent pre-determined levels of congestion in a given portion of a road network or freeway segment. (A) For the covering abstract see IRRD 873901.

  • Availability:
  • Corporate Authors:

    AA Balkema

    P.O. Box 1675
    Rotterdam,   Netherlands  BR-3000
  • Authors:
    • KUNIGAHALLI, R
    • Russell, J S
  • Publication Date: 1995

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00714381
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • ISBN: 90-5410-556-9
  • Files: ITRD
  • Created Date: Dec 27 1995 12:00AM