Pedestrians' Interaction with eHMI-equipped Autonomous Vehicles: A Bibliometric Analysis and Systematic Review

Area of Science:

• Human-Computer Interaction
• Autonomous Systems
• Road Safety

Background:

• Autonomous vehicles (AVs) require effective communication with pedestrians for safety.
• External human-machine interfaces (eHMIs) are vital for conveying AV intentions to pedestrians.
• Prior research on eHMIs has primarily focused on visual and auditory signaling.

Purpose of the Study:

• To investigate the current research landscape of eHMIs for pedestrian interactions with AVs.
• To identify key research topics, methodologies, and findings in eHMI studies.
• To highlight existing gaps and suggest future research directions in eHMI development.

Main Methods:

• Bibliometric analysis of 234 papers (2014-2023) from Web of Science.
• Systematic review of 38 selected articles focusing on pedestrian interactions.
• Analysis of research trends, methodologies, and influencing factors.

Main Results:

• Significant increase in eHMI research since 2018, focusing on crossing behavior and evaluations.
• Pedestrian behavior and eHMI perception influenced by human, vehicle, and environmental factors.
• Current research often uses virtual reality/video, lacking real-world complexity; gaps exist in multimodal systems and diverse human factor studies.

Conclusions:

• Standardization of eHMI design is crucial for effective pedestrian interaction.
• Future research should address limitations in current methodologies and explore under-researched human factors.
• Informed theoretical and practical advancements are needed for robust eHMI solutions in autonomous driving.