Empirical Evaluation of Diverse Threat Assessment Algorithms for Intersection Collision Warning

Thousands of crashes occur at intersections every year, and many of these collisions result in injuries or fatalities. Among these intersection crashes, those that occur in a Straight Crossing Path (SCP) configuration represent a sizable portion and tend to be of substantial severity. These SCP crashes are typically due to an unintentional violation by a driver. In these collisions, distraction is typically present as a causal factor. Potential countermeasures to mitigate these crashes could be aided by the development of robust warning activation algorithms that detect these potential violations in sufficient time to warn the violating driver and avert the violation and potential crash. This investigation tested several previously proposed Forward Collision Warning algorithms, along with some novel options, for use in an Intersection Collision Warning application. Simulations of the algorithms, driven by the available empirical datasets of normal and distracted intersection approaches, were created and run. Two dependent performance measures were used: 1) algorithm accuracy in appropriately identifying violation and no-violation approaches, and 2) indication of the timeliness of the warning, based on the average required constant deceleration after alarm onset. Results indicated that nuisance, missed, and late alarms are quite possible in this warning application but can be reduced by careful adjustment of algorithm warning thresholds. In addition, several of the algorithms tested performed very well during the simulations and should be considered in future development and evaluation efforts for intersection collision warning systems.

• Corporate Authors:

  Transportation Research Board

  500 Fifth Street, NW
  Washington, DC  United States  20001
• Authors:
  • Perez, Miguel A
  • Doerzaph, Zachary R
  • Neale, Vicki
• Conference:
  • Transportation Research Board 86th Annual Meeting
  • Location: Washington DC, United States
  • Date: 2007-1-21 to 2007-1-25
• Date: 2007

Language

• English

Media Info

• Media Type: CD-ROM
• Features: Figures; References; Tables;
• Pagination: 19p
• Monograph Title: TRB 86th Annual Meeting Compendium of Papers CD-ROM

Subject/Index Terms

• TRT Terms: Algorithms; Crash avoidance systems; Distraction; Driver errors; Highway safety; Intersections; Kinematics; Mathematical models; Warning systems
• Subject Areas: Highways; Safety and Human Factors; I83: Accidents and the Human Factor;

Filing Info

• Accession Number: 01044781
• Record Type: Publication
• Report/Paper Numbers: 07-2292
• Files: BTRIS, TRIS, TRB
• Created Date: Feb 8 2007 6:59PM