Facial Expressions as Indicator for Discomfort in Automated Driving
DOI: 10.1007/978-3-030-39512-4_142
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates whether facial expressions can serve as reliable indicators of discomfort in automated driving, a critical factor for public acceptance and the development of comfort-adaptive automation systems. Motivated by the EU-project MEDIATOR’s goal to create a mediating system that dynamically adjusts vehicle behavior based on driver state, the researchers aimed to identify specific facial Action Units (AUs) associated with discomfort during automated driving scenarios. The research employed a fixed-base driving simulator study involving 41 participants, divided into younger (under 40) and older (over 60) age groups. Participants experienced a prerecorded three-minute trip featuring three identical potentially uncomfortable situations: automated approaches to a slower-moving truck ahead. During these approaches, the automated vehicle closed the distance rapidly, braking at a short range of 9 meters to a minimum of 4.2 meters. Participants could continuously report perceived discomfort using a handset control. Facial data was captured via four video cameras from different perspectives and analyzed using Visage software to extract 23 AUs. The analysis focused on 98 extracted discomfort sequences, utilizing z-standardization to compare relative changes in AU activity across participants and situations. The results identified significant, consistent changes in eight AUs across all camera perspectives. During discomfort intervals, participants kept their eyes open and reduced blinking (AU43), indicating heightened visual monitoring. Simultaneously, the raising of inner brows (AU1) and upper lids (AU5) signaled surprise, while lip pressing (AU24) and stretching (AU20) indicated tension. These patterns were stable across age and gender subgroups when standardized individually. The findings suggest that specific facial muscle movements correlate strongly with reported discomfort in close-proximity automated driving scenarios. The study concludes that facial expression analysis can effectively detect discomfort, supporting the development of adaptive automation systems that adjust driving styles to prevent uneasiness or unnecessary takeovers. However, the authors note that current analyses were conducted at an aggregate level. Future research within the MEDIATOR project aims to validate these findings in diverse situations and combine facial tracking with other driver state measures to achieve reliable, real-time individual-level detection.
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-25 |
| archive | success | unpaywall | — | — | 2 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-26 |
| chunk | success | chunk | — | — | 1 | 2026-06-26 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-26 |
| enrich | failed | — | — | — | 1 | 2026-06-26 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-26 |
| verify | success | — | — | — | 1 | 2026-06-26 |
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.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Empirical Findings: physiological data
- Methodological Resource: tool software