Engaging with Highly Automated Driving: To be or Not to be in the Loop?
DOI: 10.17077/drivingassessment.1570
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 the impact of driver engagement levels on the ability to resume manual control from Highly Automated Driving (HAD) during a potential collision scenario. Motivated by the "out-of-the-loop" (OOTL) performance problem, where limited human-system interaction causes operators to lose situation awareness, the research aims to distinguish between the effects of losing sensory-motor control versus losing information processing control. The authors specifically examine whether varying degrees of engagement and task complexity (operational vs. tactical levels) influence drivers' behavioral responses when required to take over vehicle control. The experiment utilized a desktop driving simulator with 16 licensed participants in a within-subjects 3x3 repeated-measures design. The independent variables were Drive condition (manual, engaged automation, and distracted automation) and Load (no rule, congruent rule, and incongruent rule). In the manual condition, drivers controlled the vehicle. In engaged automation, drivers removed hands from controls but observed the road. In distracted automation, drivers read text on an iPad. After 60 seconds, automation disengaged, revealing a stationary obstacle. Drivers had to change lanes to avoid collision, with lane choice governed by either no rule (operational task) or color-based rules (tactical task). Dependent measures included maximum lateral/longitudinal acceleration, time to first steer, and time to lane change. Results indicated a significant main effect of Drive on maximum lateral acceleration and time to first steer. Drivers in both automation conditions exhibited significantly higher lateral accelerations and slower initial steering responses compared to manual driving, indicating more erratic and delayed control resumption. Specifically, distracted drivers responded more erratically, often reacting to the automation disengagement beep rather than the visual scene, whereas engaged drivers demonstrated more calculated maneuvers. However, there were no significant differences between engaged and distracted automation regarding the time to first steer, suggesting that the loss of physical control was the primary factor delaying response. Task load (rule complexity) did not significantly affect performance metrics, and drivers successfully adhered to tactical rules in nearly all cases. No collisions occurred. The findings suggest that the OOTL phenomenon is strongly linked to the loss of sensory-motor control rather than cognitive disengagement alone. Even brief periods of automation impaired the speed and quality of take-over maneuvers. The authors conclude that until effective strategies exist to help drivers regain situation awareness, drivers should remain in the driving loop. The study highlights the need for objective measures of take-over quality beyond reaction times and calls for further research into cognitive control loops under more complex scenarios.
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 | 2026-08-09 |
| archive | success | canonical_url | — | — | 1 | 2026-08-09 |
| extract | success | cached | — | — | 3 | 2026-08-10 |
| clean | success | clean | — | — | 1 | 2026-08-09 |
| chunk | success | chunk | — | — | 1 | 2026-08-09 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-08-09 |
| enrich | failed | — | — | — | 1 | 2026-08-09 |
| promote | success | — | — | — | 1 | 2026-08-09 |
| summarize | success | llm | qwen3.6-27b-nvidia | summ-v5 | 2 | 2026-08-10 |
| tag | success | vector_similarity | — | — | 10 | 2026-08-11 |
| verify | success | — | — | — | 2 | 2026-08-10 |
Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
- automation
- takeover transitions
- hands on hands off engagement
- automation surprise
- manual
- situational awareness
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
- Theoretical Contribution: conceptual framework, computational model