Personalized Warning Systems for Automated Driving: Adapting to Individual Driving Styles for Enhanced Takeover Performance

Spandana Sagam · 2025 · Crossref

DOI: 10.32996/jcsts.2025.7.4.104

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 how individual driving styles influence takeover performance during the transition from automated to manual control in Level 3 autonomous vehicles. The research addresses a critical safety gap: while previous studies have examined factors like secondary tasks and situational criticality, limited attention has been paid to how predisposed driving behaviors—specifically aggressive versus cautious styles—interact with warning system configurations. The authors aim to determine optimal notification strategies by analyzing how these styles affect reaction times, control resumption quality, and post-takeover stability. The methodology employed a mixed-methods approach using a high-fidelity driving simulator with motion platforms and panoramic displays. Participants were classified into aggressive or cautious driving styles through a combination of validated subjective questionnaires and objective baseline driving metrics, such as speed adherence, acceleration patterns, and following distances. The experimental design was within-subjects, exposing participants to various takeover scenarios under different warning system conditions. These conditions systematically varied timing (early, standard, late), modality (visual, auditory, haptic, multimodal), and intensity. Data collection included vehicle dynamics, eye-tracking, physiological responses (heart rate, galvanic skin response), and subjective assessments of workload and trust. Statistical analyses, including regression modeling and factorial analysis, evaluated main effects and interactions between driving styles and warning parameters. Results indicate that driving style significantly predicts takeover performance. Aggressive drivers exhibited delayed initial responses to takeover requests but demonstrated faster stabilization patterns after resuming control. In contrast, cautious drivers showed more consistent, measured reactions with greater initial stability and smoother control inputs. Warning system factors also significantly influenced performance, with timing showing particularly large effects on takeover duration. Crucially, significant interaction effects were found: aggressive drivers benefited from earlier timing parameters and more prominent sensory signals, while cautious drivers performed optimally with moderate timing and less intrusive alerts. These interaction effects remained consistent across different scenario types, suggesting stable relationships between driving styles and warning system preferences. The findings imply that personalized warning systems can substantially enhance safety during automated-to-manual transitions. By adapting timing, intensity, and modality based on detected driving patterns, vehicle manufacturers can optimize human-automation collaboration. The study supports the development of adaptive algorithms that profile driver behavior during manual operation to dynamically adjust warning parameters, thereby accommodating individual differences in attention allocation and risk perception. This approach moves beyond one-size-fits-all interfaces, offering a pathway to improve takeover reliability across diverse driver populations.

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-nvidia on 2026-08-10; verification: verified.

Topics

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

• automation
• takeover transitions
• automation surprise
• risk taking
• feedback modes
• mode 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