Rethinking External Communication of Autonomous Vehicles: Is the Field Converging, Diverging, or Stalling?

Paper Title

Rethinking External Communication of Autonomous Vehicles: Is the Field Converging, Diverging, or Stalling?

Publication Info

• Topic area: External communication interfaces for autonomous vehicles (eHMIs) and their evolution in research, industry, and regulation.
• Keywords: eHMI, autonomous vehicles, external communication, human–machine interaction, safety, standardisation, design concepts, regulation, industry alignment, multimodal interfaces.

Background and Problem

• Problem / challenge: Despite a decade of research and hundreds of studies, the field of eHMIs lacks clarity on whether it is converging on shared principles, diverging across approaches, or stalling in unresolved debates.
• Significance: Effective eHMIs are critical for safe and intuitive interactions between autonomous vehicles (AVs) and external road users, particularly pedestrians, in high-stakes traffic environments.
• Motivation and related work: Prior work has explored diverse eHMI concepts, evaluation methods, and taxonomies, but methodological inconsistencies, unresolved debates on necessity, and limited alignment with industry and regulatory practices leave the field conceptually unsettled.

Solution

• Proposed approach: A multi-dimensional analysis of 620 eHMI studies (2014–2025), combined with industry deployments and regulatory frameworks, to assess whether the field is converging, diverging, or stalling.
• Novelty:
  1. Comprehensive analysis of 620 eHMI studies, mapping the field’s evolution across key dimensions.
  2. Synthesis of convergence and divergence trends in the literature.
  3. Comparison of academic research with industry practices and regulatory developments.
  4. Evidence-based assessment of the field’s trajectory, with implications for future research and practice.
• Procedure and key techniques:
  • Structured literature review with multi-dimensional coding.
  • Analysis of industry patents, concepts, and deployments.
  • Examination of regulatory and standards documents.
  • Synthesis of trends in convergence, divergence, and stalling.

Results

• Concrete findings:
  • Pedestrians are the most studied road user group (n=424), followed by drivers (n=73) and cyclists (n=51).
  • Passenger cars dominate vehicle types studied (n=407), with limited attention to freight vehicles, delivery robots, and non-road environments.
  • Light-based eHMIs are the most common design concept (n=163), followed by displays (n=153), anthropomorphic cues (n=47), and projections (n=37).
  • Industry deployments favor simple, robust solutions like LED domes and light bars, while regulation focuses on safety-critical parameters like lighting and color.
• Advantage over baselines:
  • The study integrates academic, industry, and regulatory perspectives, revealing a progressive filtering funnel where broad research exploration narrows into practical deployments and codified standards.
• Experiments / evaluation:
  • Analysis of 620 studies, including 55 reviews and taxonomies, and industry sources from 47 companies.
  • Regulatory documents from ISO, UNECE, SAE, and national authorities.
  • Methodological focus on VR simulations, field studies, and multimodal evaluations.
• Limitations and future work:
  • Dataset restricted to English-language, peer-reviewed publications, potentially excluding non-English and grey literature.
  • Limited coverage of industry practices in regions like China.
  • Future work should prioritize longitudinal studies, ecological validity, and expert-driven consensus-building.

Summary

This paper examines the evolution of eHMI research, identifying patterns of convergence (e.g., allocentric communication, safety-first designs), divergence (e.g., minimalist vs. expressive approaches, modality debates), and signs of stalling (e.g., repetitive small-scale studies). It highlights a progressive filtering process from broad research exploration to industry deployment and regulatory standardization. The findings emphasize the need for cumulative knowledge, context-sensitive designs, and flexible but harmonized standards to support the safe and inclusive integration of AVs into public spaces.