Published November 1, 2022 | Version v1
Conference paper Open

Acceptation des véhicules entièrement autonomes par les piétons lors de situation de traversée de route : expérimentation en environnement simulé

  • 1. Luxembourg Institute of Science and Technology

Description

Pedestrians are the most vulnerable road users, especially when crossing a roadway in front of a vehicle that can inflict serious damage. Pedestrians must assess their environment and the risks involved. This will continue to be true and will raise new issues in an environment with Fully Autonomous Vehicles (FAVs). Therefore, how can the design of the external human-machine interface (eHMI) influence crossing behaviour? How receptive are pedestrians to the arrival of FAVs in the automotive landscape? To answer these questions, 39 participants were put in a pedestrian situation in front of eight simulations of autonomous cars allowing to test two different eHMIs in four situations: with or without a crosswalk on the road; the FAV stops or does not stop to let the pedestrian pass. The results show, among other things, that the responsiveness of FAVs is influenced by having dependent children under 10 years of age, nationality, and place of residence of the participants. Pedestrian profile did not appear to influence FAV receptivity. It was observed that the situations with a higher receptivity score also had a better user experience score towards the eHMI and were also the best perceived and had a shorter decision-making time. Thus, the most appreciated situation was the one where the FAV projected a crosswalk on the road when the vehicle was stopped. The situations with the lowest scores (responsiveness, user experience, situation perception and decision time) were those where the FAV did not stop and did not give any warning signal. In both cases (best and worst scores), the results show that the presence or absence of a crosswalk already on the road does not play a significant role. Finally, to ensure good acceptance of FAVs by pedestrians, we identified the participants’ needs in terms of information and training (type of information required, format and authority in charge of dissemination).

Files

3500866.3516372.pdf

Files (5.8 MB)

Name Size Download all
md5:d744787c7dc5735101ce58f2c794c3f6
5.8 MB Preview Download

Additional details

Funding

PAsCAL – Enhance driver behaviour and Public Acceptance of Connected and Autonomous vehicLes 815098
European Commission