Determining Key Parameters with Data-Assisted Analysis of Conditionally Automated Driving
DOI: 10.3390/app13116649
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 addresses the critical safety challenge of take-over requests (TORs) in conditionally automated (Level 3) vehicles, where drivers must resume control when the system reaches its functional limits. The research is motivated by the lack of consensus on which specific driving and non-driving parameters most significantly influence the quality of the take-over process, particularly when drivers are engaged in secondary tasks like using handheld devices. The authors aim to identify key parameters to inform the design of TOR user interfaces, thereby enhancing safety and comfort. The researchers utilized a dataset from a previous driving simulator study involving 36 drivers who performed 216 take-over events while playing a cognitively demanding game on a smartphone. The dataset comprised 41 attributes, including demographic data, driving parameters, and observations of handheld device handling. These attributes were categorized into pre-take-over predictors (e.g., attention), during-take-over predictors (e.g., reaction time, solution suitability), and safety performance metrics (e.g., off-road driving, braking, lateral acceleration, time to collision, and success). To determine the most influential parameters, the authors applied several interpretable machine learning models, including multiple linear regression, support vector machines, M5’, 1R, logistic regression, and J48 decision trees. The analysis revealed that maximal acceleration and the interval between the TOR and the first brake application were the primary determinants of take-over quality, with maximal acceleration enabling an 88.6% accurate prediction of collisions. The findings indicate that a braking strategy characterized by gradual, neither too hard nor too soft, deceleration executed as quickly as possible maximizes overall take-over quality. Regarding handheld device usage, the position of the device and how it was held prior to the TOR did not significantly affect take-over quality. However, handling the device during the take-over process negatively impacted driver attention, resulting in shorter attention times when drivers held their phones in only one hand. The significance of this work lies in its data-driven identification of key parameters that can guide the optimization of TOR user interfaces. By understanding which factors most impact safety, designers can adjust interface modalities and timing to mitigate risks. The study suggests that automatic gradual braking maneuvers could be considered as a countermeasure to immediate full take-overs, potentially improving safety outcomes. Furthermore, the results highlight the specific risks associated with single-handed device handling during critical transitions, providing actionable insights for future research and vehicle system design in the era of conditional automation.
Key finding
Maximal acceleration and the time interval between the take-over request and the first brake application are the primary parameters determining the quality and safety of the take-over process.
Methodology
dataset
Sample size: 216
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 topic_sweep_doaj on 2026-06-01.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | — | — | — | 1 | 2026-06-01 |
| archive | success | openalex | — | — | 5 | 2026-06-06 |
| 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 | — | — | — | 1 | 2026-06-01 |
| 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.
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: behavioral performance data
- Methodological Resource: measurement protocol
- Theoretical Contribution: conceptual framework