• DocumentCode
    679213
  • Title

    Model of fleet-wide safety benefits of Lane Departure Warning systems

  • Author

    Gorman, Thomas I. ; Kusano, Kristofer D. ; Gabler, Hampton C.

  • Author_Institution
    Virginia Tech, Blacksburg, VA, USA
  • fYear
    2013
  • fDate
    6-9 Oct. 2013
  • Firstpage
    372
  • Lastpage
    377
  • Abstract
    This study examines the potential effectiveness of Lane Departure Warning (LDW) systems, an active safety system that aims to prevent or mitigate unintended departures from the lane while driving. Real-world departure crashes were extracted from a nationally representative sample of collisions in the U.S. from 2007 to 2011. The 2,848 collisions corresponded to 741,646 crashes. Because driver characteristics, e.g. steering input and reaction times, are unknown in retrospective crash databases, multiple simulations were performed for each crash, each with different driver characteristics. In addition, we developed a statistical framework for estimating initial vehicle conditions, i.e. departure speed, departure angle, road radius of curvature, and shoulder width. Trajectories were simulated to estimate the expected number of crashes and injuries that could be prevented by LDW. The expected reduction in road departure crashes due to LDW is estimated to be 30.3%. The expected reduction in injuries that are caused by LDW applicable crashes is estimated to be 25.8%.
  • Keywords
    computer vision; driver information systems; road safety; LDW; US; active safety system; departure angle; departure speed; driver characteristics; fleet-wide safety benefits; lane departure warning systems; nationally representative collisions sample; reaction times; retrospective crash databases; road curvature radius; shoulder width; steering input; Computer crashes; Injuries; Mathematical model; Roads; Trajectory; Vehicle crash testing; Vehicles;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Transportation Systems - (ITSC), 2013 16th International IEEE Conference on
  • Conference_Location
    The Hague
  • Type

    conf

  • DOI
    10.1109/ITSC.2013.6728260
  • Filename
    6728260