Development of a Monitoring System for Driver Readiness in Prolonged Automated Driving
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This study addresses the critical safety challenge of driver fatigue during prolonged automated driving, specifically within SAE Level 3 conditional automation. As automated vehicles remove drivers from active control, passive fatigue can develop, compromising their readiness to take over control during system failures. The research aims to develop a non-intrusive, robust monitoring system that detects varying levels of driver fatigue by integrating postural and behavioral data. The researchers conducted a driving simulator experiment involving twenty licensed drivers. Participants engaged in a 45-minute simulated drive on a highway scenario, divided into three 15-minute sessions with breaks to mitigate simulation sickness. The study utilized a 3DOF motion-base simulator equipped with SAE Level 3 automation capabilities. Data collection employed two cameras: a GoPro HERO10 facing the driver to capture facial features, and a Logitech webcam positioned behind the driver to record postural metrics. Key features extracted via computer vision included eye aspect ratio (EAR), mouth opening ratio (MOR), percentage of eye closure (PERCLOS), head position, head-to-headrest distance, and hand-to-steering wheel distance. The study evaluated various machine learning algorithms, including Random Forest, Support Vector Machines, and Decision Trees, to classify fatigue levels based on these multimodal inputs. The primary finding was that postural data served as the most critical factor in detecting driver fatigue, outperforming facial features alone. The Random Forest algorithm demonstrated the highest performance among the tested models, achieving an accuracy of 0.97 in detecting fatigue levels. The study confirmed that combining postural data with physical measures such as EAR, MOR, and PERCLOS provided a highly effective and accurate method for identifying driver fatigue. This integrated approach proved superior to models relying on single-modal data, addressing previous research gaps regarding the generalizability and robustness of fatigue detection systems. The significance of this work lies in its contribution to the safety infrastructure of automated vehicles. By validating a non-intrusive monitoring system that leverages readily available camera technology, the study offers a practical solution for real-time driver state assessment. The high accuracy of the Random Forest model suggests that such systems can reliably alert drivers when their readiness deteriorates, thereby reducing the risk of crashes caused by delayed or failed takeovers. This research supports the broader adoption of automated driving technologies by providing a mechanism to mitigate the specific risks associated with driver disengagement and fatigue.
Key finding
Postural data is the most critical factor for detecting driver fatigue, and a random forest algorithm combining postural and facial features achieved an accuracy of 0.97.
Methodology
simulator
Sample size: 20
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 bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 19 | 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.
- drowsiness detection algorithms
- situational awareness
- drowsiness
- distraction detection algorithms
- vigilance
- truck driver fatigue
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, validation psychometrics