Naturalistic Driving Data
UMTRI has developed large driver-vehicle databases since the mid-1990s. These not only answer specific questions about the systems being tested (e.g., adaptive cruise control), but also contain a rich resource of general driving data, representing how people drive in the real world. Current (2006) resources represent approximately 1 million vehicle miles traveled, and much of this data is highly detailed, containing hundreds of signals measured from the subject vehicle, radar measurements of nearby vehicles, and video data of both the forward scene and the driver\'s activities. UMTRI also has an extensive SAVME database of highly detailed vehicle motions, with over 35,000 vehicle trajectories analyzed and recorded based on video motion capture. The next-generation SAVME system will record even more detailed data of the same type, particularly related to vehicle kinematics at intersections.
Recent PublicationsSEE MORE
John M. Sullivan, Shan Bao, Roy Goudy, Heather Konet
journal article IN: Accident Analysis and Prevention, volume 74, January 2015, pages 1-7
The purpose of this study was to determine whether a driver's use of turn signals is sufficiently reliable to forecast a vehicle...
Timothy J. Gordon, Lidia P. Kostyniuk, Paul E. Green, Michelle A. Barnes, Daniel F. Blower, Scott E. Bogard, Adam D. Blankespoor, David J. LeBlanc, B. R. Cannon, S. B. McLaughlin
report SHRP2 Report S2-S01C-RW-1
This report documents and presents the results of a preliminary study into the use and validation of crash surrogates, to be obtained...
David J. LeBlanc, Michael Sivak, Scott Bogard.
Fuel consumption rates were studied from a naturalistic driving data set employing a fleet of identical passenger vehicles with...
Tim Gordon, Adam Blankespoor, Michelle Barnes, Dan Blower, Paul Green, Lidia Kostyniuk.
conference paper 09-0326
The aim of this work is to define and evaluate a -yaw rate error- (YRE) derived from naturalistic driving data to quantify driver...
Matthew P. Reed.
IN: Sponsored by: University of Michigan Industry Affiliation Program for Human Factors in Transportation Safety.
Vehicle turn trajectories from a naturalistic driving database were modeled using Bezier curves. A-pillar geometry from 56 vehicles was...