Übernahmeaufforderungen beim hochautomatisierten Fahren mit fahrfremden Tätigkeiten – welche Modalitäten sind geeignet? Takeover requests at conditional automated driving with non-driving related activities—which are suitable modalities?

Müller, Andreas Lars; Abendroth, Bettina · 2022 · Crossref

DOI: 10.1007/s41449-021-00295-2

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 suitability of different modalities for Takeover Requests (TORs) in conditional automated driving (SAE Level 3), specifically when drivers are engaged in non-driving related tasks (NDRTs). The research addresses two primary questions: whether multimodal TORs yield faster reaction times than unimodal ones, and how the mental workload of various NDRTs impacts takeover performance. The authors conducted two independent driving simulator studies with a total of 126 participants. Study 1 evaluated three TOR interface variants: visual (LED light strips), vibrotactile (a custom vibration mat with 21 actuators), and multimodal (combining visual, vibrotactile, and acoustic signals). Participants performed NDRTs such as smartphone use or a 2-Back test while driving. Study 2 focused on the impact of specific NDRTs—reading text, listening to audiobooks, watching videos, texting, and observing the road (reference)—on takeover duration using the multimodal TOR. Mental workload was measured using a Detection Response Task secondary paradigm. Results from Study 1 demonstrated that multimodal TORs significantly reduced reaction times compared to both visual (mean: 1.89 s) and vibrotactile (mean: 1.39 s) variants, with the multimodal average being 1.17 s. Subjectively, multimodal and visual TORs were rated as more urgent and helpful than vibrotactile ones, though no significant differences were found regarding disturbance or comfort. Study 2 revealed that reaction times correlated strongly with the mental demands of the NDRT. Reading text resulted in the longest average reaction time (1.64 s), followed by watching videos (1.48 s) and texting (1.49 s). Listening to audiobooks (1.10 s) and observing the road (1.11 s) yielded the shortest reaction times. Statistical analysis confirmed that takeover duration increased significantly with higher mental workload. The findings imply that multimodal TORs are superior for ensuring rapid driver response in automated vehicles. Furthermore, the study highlights that the type of NDRT significantly affects takeover capability, with cognitively demanding tasks like reading text posing greater risks than passive tasks like listening to audio. These results provide practical guidelines for designing human-machine interfaces in automated driving, emphasizing the need for multimodal alerts and careful consideration of driver engagement levels during automation.

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.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• takeover transitions
• automation
• feedback modes
• automation surprise
• mode awareness
• haptic feedback

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