A Systematic Review of VR-Based Studies on Pedestrian Interaction with eHMI-Equipped Automated Vehicles: From Simplified Paradigms to Social Contexts

Abstract

With the advancement of automated vehicles (AVs), ensuring safe interaction between pedestrians and AVs become a critical challenge. External human–machine interfaces (eHMIs) have emerged as a solution for communicating vehicle intentions to pedestrians. In recent years, virtual reality (VR) has been increasingly used in this research area because of its safety, cost-effectiveness, and high level of experimental control. To synthesise current research trends and identify emerging gaps, this paper systematically reviews 22 VR-based studies on pedestrian interaction with eHMI-equipped AVs published between 2009 and 2025. The results show that VR-based pedestrian-AV research has gradually shifted from simplified, one-to-one interaction paradigms toward more complex experimental designs that incorporate complex traffic environments, diverse participant characteristics, various vehicle settings and eHMI designs. The review identifies an emerging focus on the social context of multi-pedestrian scenarios and its influence on pedestrian crossing behaviour, highlighting VR’s potential for studying such behavioural interactions.