Effects of a frontal brake light on pedestrians' eye-movements and reaction times in the perception of vehicle deceleration
Tue-H5-Talk 5-4901
Presented by: Daniel Eisele
Over the last years, there has been a lively debate on whether (automated) vehicles should be equipped with novel external human-machine interfaces (eHMIs) in order to facilitate communication with nearby vulnerable road users. One simple approach is a frontal brake light (FBL), which indicates to road users in front of the vehicle that the vehicle is decelerating. It has been established that FBLs can facilitate the identification of decelerations and thereby efficiently communicate potentially important information about the vehicle’s behavior. It seems plausible that exposure to FBLs alters eye-movements of observers trying to detect decelerations (e.g., visual attention is focused on the area where one expects an FBL to activate). As a possible consequence, the detection of decelerations in vehicles without an FBL may be impaired.
To investigate possible changes in eye movements caused by the presence of FBLs in traffic, an eye-tracking experiment was conducted. Twenty-nine participants were shown 138 videos of approaching vehicles from the perspective of a pedestrian standing at the curb. The vehicles either started decelerating (at three different distances from the participants) or maintained their speed. Over the course of the experiment, the quota of FBL-equipped vehicles in simulated traffic was increased (from 0% to 83%). The participants’ task was to indicate the moment they perceived the deceleration.
First analyzes show large effects of the presence of an FBL on reaction times. Eye-movement behavior differed between the experimental conditions. More detailed results will be reported. Implications for the development of eHMIs will be discussed.
To investigate possible changes in eye movements caused by the presence of FBLs in traffic, an eye-tracking experiment was conducted. Twenty-nine participants were shown 138 videos of approaching vehicles from the perspective of a pedestrian standing at the curb. The vehicles either started decelerating (at three different distances from the participants) or maintained their speed. Over the course of the experiment, the quota of FBL-equipped vehicles in simulated traffic was increased (from 0% to 83%). The participants’ task was to indicate the moment they perceived the deceleration.
First analyzes show large effects of the presence of an FBL on reaction times. Eye-movement behavior differed between the experimental conditions. More detailed results will be reported. Implications for the development of eHMIs will be discussed.
Keywords: human-robot interaction, traffic, pedestrian, communication, eye-tracking