A Contextual Multimodal System for Increasing Situation Awareness and Takeover Quality in Conditionally Automated Driving
DOI: 10.1109/access.2023.3236814
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 introduces AdVitam, an adaptive multimodal system designed to enhance driver situation awareness (SA) and takeover quality in conditionally automated vehicles (SAE Level 3). The research addresses the safety and trust challenges associated with the shift in driving responsibility from humans to machines, particularly when drivers engage in non-driving-related tasks. The system aims to dynamically adapt human-vehicle interaction based on the driver’s physiological state and the driving environment to maintain SA and optimize the driver’s readiness to resume control. The AdVitam system comprises three interconnected modules. The Driver State module uses machine learning to predict four risk factors—fatigue, mental workload, affective state, and SA—from continuous physiological signals (ECG, EDA, and respiration). The Supervision module maintains SA by conveying context-related information via ambient lights, haptic seat vibrations, and a mobile application, based on environmental status and driver state. The Intervention module employs a machine learning model to select the optimal multimodal combination (haptic, auditory, visual) for Take-Over Requests (TORs) to minimize reaction time and steering errors. The system was evaluated in a preliminary user study with 35 participants in a fixed-base driving simulator across rural and urban environments, with the Supervision and Intervention modules manipulated as between-subject factors. Results indicate that conveying environmental status through multimodal interfaces significantly increased drivers’ situation awareness, evidenced by better identification of potential environmental problems, and improved trust in the automated vehicle. The Driver State module demonstrated consistent predictions aligned with experimental manipulations. However, the system did not yield positive outcomes regarding takeover quality metrics. The study validates the effectiveness of the Supervision module for enhancing SA and trust but highlights limitations in optimizing the actual takeover performance through the proposed Intervention strategies. The significance of this work lies in its contribution to the design of adaptive human-vehicle interaction systems for automated driving. By demonstrating that multimodal supervision can enhance SA and trust, the findings suggest that such systems could improve the acceptance and safety of conditionally automated vehicles. The study underscores the importance of maintaining driver awareness during automation phases, although it also reveals the complexity of optimizing takeover quality, indicating a need for further refinement in intervention strategies.
Key finding
Multimodal contextual information increased drivers' situation awareness and trust, but did not improve takeover quality.
Methodology
simulator
Sample size: 35
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. Discovered via scout_discovery on 2026-05-08.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | partial | scout | — | — | 2 | 2026-05-08 |
| archive | success | unpaywall | — | — | 1 | 2026-06-04 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-04 |
| chunk | success | chunk | — | — | 1 | 2026-06-04 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-04 |
| enrich | success | semantic_scholar | — | — | 2 | 2026-06-04 |
| promote | success | — | — | — | 1 | 2026-06-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- situational awareness
- automation
- mode awareness
- automation surprise
- takeover transitions
- distraction detection algorithms
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).
- Theoretical Contribution: conceptual framework, theory or model, computational model