Analysis of Driver Judgment and Reaction by Different Levels of Visual Information on eHMI

Abstract

As autonomous vehicles make their debut on public roads, communicating with other drivers via an external human-machine interface (eHMI) has become crucial for road safety. Our work explores required visual information levels by observing participants’ judgment time in the driver’s seat and by measuring the ability metrics of eHMI implemented on autonomous vehicles (AVs) passing by. We evaluated about 10 eHMI design alternatives, and measured the time from perceiving the rear-facing eHMI of an AV passing by to the moment the participants decide how to drive using eye tracking glasses. Also, we surveyed subjective evaluations of how participants feel about seeing the eHMIs. The results show that using text reduces the time to drivers’ judgment, and fast flashing effects and the color red effectively draw attention and give warning. The results suggest that it is essential for future AVs to use text and practical visual elements when other drivers on the road require quick reactions. © 2021, Springer Nature Switzerland AG.