Influence of an Additional Pedestrian and an External Human-Machine Interface on one's Crossing Behavior in front of an Automated Vehicle in a Shared Space Environment

Kurzfassung

The introduction of automated vehicles (AVs) into public road traffic is expected to bring numerous benefits but also poses new challenges for the existing traffic system. Among these challenges, the interaction between AVs and other road users (RU) plays a vital role. Especially for pedestrians or cyclists, who are vulnerable road users, safe and predictable interaction with AVs becomes crucial. Although the crossing behavior of pedestrians in front of an AV has been studied previously, the investigation has not explored conflicting situations involving the presence of additional pedestrians. Hence, the question remains as to how the presence of other pedestrians influences an individual’s decision to cross in front of an AV. Further, does the presentation of the AV’s intentions via an external human-machine-interface (eHMI) influence the crossing behavior, and do entrainment effects exist when other pedestrians are present in the same situation? Therefore, this experimental online study (N = 136) aimed to explore these questions using video-animated scenarios combined with a questionnaire. The results align with previous research that says that willingness to cross and perceived safety are influenced by information from the eHMI and vehicle kinematics, but they also revealed that they are further impacted by pedestrian behavior. The significant impact of pedestrian behavior on the willingness to cross and perceived safety illustrates the importance of acknowledging the influence of other actors on one’s own decision-making process. It highlights the relevance of understanding the underlying factors that influence a person’s interaction with automated vehicles. These findings are particularly important for the recognition, learning, and decision-making processes of automated vehicles itself and can, therefore, also have a corresponding significance for the implementation of highly developed eHMIs as a means of interaction.