Autonomous driving: investigating the feasibility of car-driver handover assistance

Walch, Marcel; Lange, Kristin; Baumann, Martin; Weber, Michael · 2015 · OpenAlex

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 paper investigates the feasibility and design of car-driver handover assistance in highly automated vehicles, addressing the critical transition from autonomous to manual control when system boundaries are reached. The authors argue that while fully autonomous vehicles are unlikely to replace manual driving immediately, highly automated systems will require mechanisms for drivers to take over control in critical situations. The research focuses on designing a handover assistant that enables drivers to regain control safely and comfortably, even when they are “out of the loop” due to secondary tasks. The study aims to evaluate specific warning strategies and handover procedures to mitigate the risks associated with this transition. To test these concepts, the researchers developed a handover assistant prototype and conducted an in-lab user study using a driving simulator. The study involved 30 participants who engaged in a realistic distractor task (watching videos) while the vehicle operated in autonomous mode. The experimental scenario involved the vehicle approaching a screen of fog, a condition simulating sensor failure, which triggered a handover request. The study employed a 3x3 repeated-measures design, varying three warning conditions: an alert combined with a take-over request (ALERT&TOR), a take-over request with a 4-second time budget (TOR4SECONDS), and a take-over request with a 6-second time budget (TOR6SECONDS). Additionally, three post-handover situations were tested: no hazard, a sharp curve, and a broken-down car. The handover procedure implemented was “immediate,” where control shifted as soon as the driver grasped the steering wheel. Multimodal warnings, consisting of visual cues and auditory earcons followed by verbal messages, were used to alert participants. The results indicated that the type of warning significantly influenced take-over time. Participants took significantly longer to grasp the steering wheel in the ALERT&TOR condition compared to the TOR4SECONDS and TOR6SECONDS conditions, likely due to the additional processing time required for the initial alert. However, the time budget itself (4 vs. 6 seconds) did not significantly affect take-over times. Regarding braking behavior, participants braked more frequently and earlier in hazardous situations (broken-down car or sharp curve) compared to the no-hazard scenario, regardless of the warning type. Perceived comfort was highest in the no-hazard situation. Qualitative data revealed that participants generally felt they could manage the situation safely, though some expressed stress during the take-over. The study also highlighted that distracted drivers were capable of taking over control within the provided time budgets, though individual variability was high. The significance of this work lies in its contribution to the design of human-machine interfaces for autonomous vehicles. The findings suggest that multimodal warnings are a promising strategy for compensating for system boundaries, but the design must balance information clarity with response speed. The authors conclude that immediate handovers are feasible but require careful consideration of warning design to ensure driver comfort and safety. The study provides recommendations for future handover assistants, emphasizing the need for realistic evaluation settings and the importance of understanding driver behavior during the transition from automation to manual control. These insights are crucial for developing safe and effective automated driving systems that can reliably hand over control to human drivers when necessary.

Key finding

Multimodal warnings that combine an alert explaining the situation with a take-over request result in longer take-over times than take-over requests alone, but remain a viable strategy for ensuring safe transitions from autonomous to manual driving.

Methodology

simulator

Sample size: 30

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.

StageOutcomeToolModelPromptAttemptsCompleted
discover success 1 2026-05-07
archive success canonical_url 7 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 2026-05-07
promote success 1 2026-05-07
summarize success llm qwen3.6-27b-prismaquant summ-v5 2 2026-06-10
tag success vector_similarity 15 2026-06-11
verify partial 2 2026-06-10

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified_with_issues.

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).