Skip Navigation

Mid-America Transportation Center

Modeling Driver Behavior and Driver Aggressiveness Using Biobehavioral Methods


University

University of Kansas

Principal Investigator
Alexandra Kondyli
PI Contact Information
akondyli@ku.edu
Funding Source(s) and Amounts Provided
USDOT: $84,128
Additional Funding: $84,389
Total Project Cost
$ 168,517
Agency ID or Contract Number
69A3551747107
Start Date
07/15/2017
In Progress
End Date
12/04/2018
Brief Description of Research Project
It is well known that driver inattention and human error are the primary causes of traffic accidents. In addition, existing driver behavioral modeling algorithms (e.g., car-following, lane changing) assume that driver variability is expressed through various distributions and random number generators. What constitutes aggressive driving, and which are the actions of aggressive drivers that negatively affect safety and traffic instability, are some of the topics that have not been studied thoroughly. At the same time, significant work has been done in the field of cognitive science and psychology, with emphasis in understanding, modeling, and predicting drivers’ intended actions.
The goal of this research is to investigate the linkage between different driver profiles with both traffic stability and the probability of being involved in risk-taking behaviors, borrowing concepts from the fields of cognitive science and psychology. Participants with different driving habits and levels of aggressiveness will be invited to participate in driving simulator experiments, where they will be asked to drive under different geometric, control, and traffic scenarios, that may additionally vary on the level of moral decision making involved. Various metrics related to drivers’ reaction times, gap acceptance, car-following, and lane changing activity will be measured through the driving simulator experiments. Additional behavioral and psychophysical measures will be collected through electroencephalogram recordings (EEG) during the simulator experiments, and through questionnaires.
These data will result in the identification of measurable behavioral parameters and their inter-driver heterogeneity. It is expected that these parameters will be used in subsequent projects to refine or develop enhanced driver behavior models that account for both safety and traffic instabilities.
Describe Implementation of Research Outcomes
Impacts/Benefits of Implementation
Web Links