| Title: |
Preliminary comparisons between driver responses to unexpected pedestrian crossings when using a head-mounted display versus 300 degree wrap around screens in a fixed-base driving simulator
|
|
Authors:
|
Danielle Filio; Diego Gonzalez; Lynn Dony; Ryan Toxopeus; Michele Oliver
|
|
Session:
|
Poster Session - Colony Ballroom, 16 Oct 2016, 5:30 PM - 7:00 PM
|
|
|
Background & Objectives Road traffic has increased dramatically as a result of people using vehicles directly for their occupation or indirectly as they drive to work. It is important to understand how people respond to road hazards in order to understand how to improve safety. Driving simulators provide a safe way to assess driver responses to hazards. Wrap-around (WA) or single screens have been used in most driving simulators but recent advances in Head Mounted Display (HMD) technology may provide a space and cost saving alternative. Though the HMD may conserve space and reduce cost, it may also limit driver field of view (FOV) thus impacting hazard perception. The objective of this preliminary study was to compare driver reactions to unexpected pedestrian crossings between a HMD and WA screens in a fixed-based driving simulator. Methods Prior to data collection, participants (n=8; 20.3±2.2 years) completed familiarization drives in a fixed-based, full-scale driving simulator (Oktal, Toulouse, France) using two display modalities (six WA screens with 300° FOV and an Oculus Rift™ DK2 HMD with 100° FOV). In each of the two approximately 10 minute data collection drives, the display modality was randomized, and a pedestrian would unexpectedly cross the road forcing the driver to react. Simulator data were collected using Oktal SCANeR™ Studio and analyzed using custom MATLAB™ code (The MathWorks Inc., MA, USA). Results ANOVAs (p≤0.05) revealed no significant differences between the HMD and WA conditions for all variables (velocity, standard deviation of lane gap, brake pressure loading, time to maximum brake pressure and reaction time). While the power to detect significant differences was relatively low, for most of the variables, approximately 25 subjects would be required with the exception of brake pressure loading where only 7 subjects were needed. Differences between means of the HMD and WA for all variables was very small. Conclusions While research is ongoing, these initial results suggest that there are minimal differences in driver behaviour between HMDs and WA screens in response to pedestrian hazards. Work is ongoing to definitively determine if HMDs can be used over WA displays for the assessment of driver hazard responses.
|