Presentation
Effects of Inconsistency between Perceived Implicit Cues and Explicit (eHMI) Cues of Autonomous Vehicles on Street-crossing Decisions
SessionWednesday Poster Session
DescriptionPedestrians typically rely on cues that are implicit (e.g., vehicle deceleration) or explicit (e.g., drivers’ gestures) to make street-crossing decisions. However, changes in speed (e.g., deceleration) are difficult to perceive, and driverless autonomous vehicles (AVs) do not have drivers to gesture intentions. External human–machine interfaces (eHMIs) have been proposed to communicate AVs’ intentions. However, a pedestrian’s perception of a vehicle’s deceleration may not match the vehicle’s actual motion, and effects of this mismatch on street-crossing decisions and trust in AVs are unknown. The current study used virtual reality displays to examine whether mismatches between perception of an AV’s deceleration and explicit eHMI cues affect street-crossing decisions, compliance with the cues, trust in the AV, and response time (RT). When perceived implicit cues and explicit cues were mismatched, people relied more on their perception of the vehicle’s deceleration than on eHMI cues alone. Compliance and trust in the AV system were highest when the eHMI cue matched participants’ perceptions of deceleration. RTs depended on both the participants’ perceptions and vehicle speed. The success of eHMIs depends not only on whether they communicate AV intention, but also on whether they are consistent with the pedestrian’s perception of AV cues.
Event Type
Poster
TimeWednesday, October 21st5:30pm - 6:30pm PDT
Location
Similar Presentations