External communication of automated shuttles: Results, experiences, and lessons learned from three European long-term research projects

Mirnig, Alexander G.; Gärtner, Magdalena; Fröhlich, Peter; Wallner, Vivien; Dahlman, Anna Sjörs; Anund, Anna; Pokorny, Petr; Hagenzieker, Marjan; Bjørnskau, Torkel; Aasvik, Ole; Demir, Cansu; Sypniewski, Jakub · 2022 · Crossref

DOI: 10.3389/frobt.2022.949135

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the communication gap between automated shuttles and other road users during the transition to fully automated mobility. As automated vehicles lack human drivers who traditionally communicate intent through gestures and eye contact, there is a heightened need for external human-machine interfaces (eHMI) to ensure safety and trust in mixed traffic environments. The study synthesizes results from seven studies (three preparatory and four main) conducted across three European long-term research projects: Austria’s *auto.Bus—Seestadt* and *Digibus Austria*, and the Horizon 2020 project *Drive2TheFuture*. The research aims to evaluate various eHMI designs for communicating with both motorized and non-motorized road users in critical scenarios such as crossings, junctions, and boarding. The methodology involved a multi-phase approach beginning with an expert workshop to identify critical interaction scenarios, including unregulated junctions, regulated junctions, crossings, passing, and boarding. Initial eHMI concepts—morphing arrows, icons, and LED bars—were evaluated via online questionnaires using 3D animations. Based on these findings, iterated designs were developed, including animated countdowns, simplified icons/arrows, and animated bars, which were tested in a second online questionnaire using video overlays and an initial field trial on a test track. The main studies included field trials in Austria and Norway, as well as a co-simulation study in Sweden, focusing on specific contexts like reducing dangerous overtaking and supporting pedestrian communication during docking. Participants rated the interfaces on safety, clarity of intent, and recognizability using Likert scales and System Usability Scale items. The findings indicate that eHMIs can effectively compensate for the lack of human driver communication, particularly in ambiguous situations. The preparatory studies revealed that initial designs like morphing arrows and icons were perceived as suitable for communicating driving intentions, while LED bars were less positively received. Iterated designs, specifically animated countdowns, proved effective in resolving "deadlock situations" at intersections by clearly signaling the shuttle’s departure time, thereby addressing the vehicle’s extended start interval. The field and simulation studies demonstrated that specific eHMI configurations could reduce dangerous overtaking behaviors and improve pedestrian confidence during boarding and docking operations. However, the research also highlighted challenges regarding visibility and readability over distance, leading to the refinement of designs to be simpler and more intuitive. The significance of this work lies in providing evidence-based design recommendations for external communication in automated public transport. By pooling data from multiple European projects, the study offers a comprehensive overview of what works in real-world and simulated traffic contexts. The results suggest that while eHMIs are valuable tools for increasing trust and safety during the transition phase, their design must be carefully tailored to specific scenarios to avoid confusion. The paper concludes that clear, universally understandable indicators are essential for facilitating smooth interactions between automated shuttles and other road users, contributing to the broader goal of integrating automated mobility safely into existing infrastructure.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• ehmi external hmi
• vru facing ehmi