How Can Autonomous Vehicles Convey Emotions to Pedestrians? A Review of Emotionally Expressive Non-Humanoid Robots

Wang, Yiyuan; Hespanhol, Luke; Tomitsch, Martin · 2021 · OpenAlex-citations

DOI: 10.3390/mti5120084

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 challenge of facilitating safe and socially acceptable interactions between autonomous vehicles (AVs) and pedestrians. The absence of human drivers removes traditional communication cues like eye contact and gestures, creating an "interaction void." While current research focuses on external human-machine interfaces (eHMIs) that convey intent and awareness through pragmatic signals like LED patterns or projections, the role of emotional expression remains largely unexplored. The authors argue that incorporating affective interfaces can enhance social acceptance, evoke empathy, and regulate traffic behavior by shifting the perception of AVs from mindless machines to social actors. To establish a foundation for affective AV-pedestrian interfaces, the study reviews how non-humanoid robots—entities similar to AVs in their lack of anthropomorphic features—convey emotions. The authors conducted a systematic review of 25 articles published between 2011 and 2021. They searched databases including IEEE Xplore, ACM Digital Library, and ScienceDirect using keywords such as "emotion," "robot," and "non-humanoid." The selection criteria required articles to involve non-humanoid robots, design specific emotional expressions, and evaluate these expressions through empirical user studies. The review analyzed the articles based on four research questions: the emotions expressed, the modalities used for display, the evaluation measures employed, and user perceptions of these expressions. The findings reveal that non-humanoid robots primarily utilize visual, auditory, and haptic modalities to encode emotions. Visual modalities were the most prevalent, with movement used in 22 articles and colored lights in seven. Movement patterns were often derived from biological motion studies, conceptual metaphors (e.g., "joy is up"), or anthropomorphic behaviors adapted for non-humanoid forms. Facial expressions, including animated eyes, were also employed in some studies. Auditory cues, such as non-linguistic utterances and music, appeared in 32% of the articles, while haptic feedback was less common. Regarding emotion models, 76% of the studies used categorical models, predominantly Ekman’s six basic emotions, while 36% utilized dimensional models like Russell’s circumplex model or Mehrabian’s PAD model. A small subset used emotional personas to define robot personalities. The significance of this work lies in its translation of design paradigms from social robotics to autonomous vehicle interfaces. By demonstrating that non-humanoid robots can effectively convey emotions through diverse modalities without relying on humanoid features, the paper provides a set of considerations for designing affective eHMIs for AVs. The authors conclude that integrating emotional expressiveness into AVs can help overcome psychological barriers, increase public trust, and improve the overall social acceptance of autonomous technology in urban environments. This review highlights avenues for future research into how affective cues can complement pragmatic signals to create more intuitive and socially robust AV-pedestrian interactions.

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