Self-Driving Vehicles and Pedestrian Interaction: Does an External Human-Machine Interface Mitigate the Threat of a Tinted Windshield or a Distracted Driver?

ABSTRACT

With self-driving vehicles (SDVs), pedestrians lose the possibility of making eye contact with an attentive driver. This study investigated whether an external human-machine interface (eHMI) displaying the automated driving mode (a. without eHMI vs. b. with eHMI) affects how pedestrians respond to different driver’s states: (1) attentive driver, (2) tinted windshield, (3) distracted driver (within-subject design). At a test site, N = 65 pedestrians crossed a pedestrian crossing while a Wizard-of-Oz SDV approached. We assessed perceived safety and crossing onset times after each trial. Findings reveal that without an eHMI, pedestrians felt significantly less safe if the windshield was tinted or the driver was distracted as compared to an attentive driver. With an eHMI, pedestrians did not differ in perceived safety with regard to the driver’s state. We observed no significant differences in pedestrians’ crossing onset times. We conclude that an eHMI helps pedestrians to not consider the driver’s state.

Acknowledgments

We would like to thank Daniel Alius, Juergen Arnold, Michael Boehringer, Andreas Calzetta, Edwin Danner, Tom Ehrhart, Joseph Gaulin, Miriam Gieselmann, Sonja Hildebrandt, Ulrich Hipp, Ralf Krause, Benno Loeffler, Bernhard Noack, Johanna Ohnmacht, Frank Ruff, and Carmen Zauner for their help with the study as well as all study participants for their time and feedback.

Data availability statement

Data is not available due to legal restrictions.

Disclosure of potential conflict of interest

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. There is no potential conflict of interest.

Additional information

Notes on contributors

Stefanie M. Faas

Stefanie Martina Faas is a current PhD candidate at Ulm University. Her research examines in which ways pedestrians’ interactions with self-driving vehicles can be facilitated via an external human-machine interface. At the time the majority of the work was done, she was employed at Mercedes-Benz AG.

Vanessa Stange

Vanessa Stange studied Psychology at Göttingen University receiving her MSc in 2018. She is currently a PhD candidate at the Department of Traffic and Engineering Psychology at Technische Universität Braunschweig. Her research focuses on driver and passenger interactions with highly automated vehicles in mixed traffic.

Martin Baumann

Martin Baumann is full professor at Ulm University and chair of the Human Factors department since 2014. His main research interests are cognitive and emotional processes underlying the interaction of humans with intelligent systems, concepts of cooperative human-machine interaction in different domains, mainly traffic, and human-robot interaction.