Construction of “Driving Etiquette” database

The “Driving Etiquette” concept aims to establish statistics from large quantity of naturalistic driving data, and use the finding to ensure autonomous vehicles drive “like safe human drivers”. The motivation is from the observation that Google cars have been involved in 15 crashes over its first 1.4 million miles. This is about four times more frequent than human drivers’ crash statistics. The project team hypothesizes that if highly autonomous vehicles behave more like “normal” human drivers, they can operate more seamlessly in the traffic, and the crash rate may improve. “Learning etiquette” involves collecting a large amount of video data, and use the data to train algorithms to learn about “what is appropriate”.


  • English


  • Status: Active
  • Contract Numbers:


  • Sponsor Organizations:

    Department of Transportation

    Office of the Secretary
    1200 New Jersey Avenue, S.E.
    Washington, DC    20590

    Office of the Assistant Secretary for Research and Technology

    University Transportation Centers Program
    Department of Transportation
    Washington, DC  United States  20590
  • Managing Organizations:

    Center for Connected and Automated Transportation

    University of Michigan, Ann Arbor
    Ann Arbor, MI  United States  48109
  • Performing Organizations:

    University of Michigan, Ann Arbor

    Department of Civil and Environmental Engineering
    2350 Hayward
    Ann Arbor, Michigan  United States  48109-2125
  • Principal Investigators:

    Peng, Huei

  • Start Date: 20170901
  • Expected Completion Date: 20191231
  • Actual Completion Date: 0
  • USDOT Program: University Transportation Centers Program

Subject/Index Terms

Filing Info

  • Accession Number: 01645445
  • Record Type: Research project
  • Source Agency: Center for Connected and Automated Transportation
  • Contract Numbers: 69A3551747105
  • Files: UTC, RiP
  • Created Date: Sep 5 2017 6:33PM