Semi-automated versus highly automated driving in critical situations caused by automation failures

Strand, Niklas; Nilsson, Josef; Karlsson, MariAnne; Nilsson, Lena · 2014 · OpenAlex-citations

DOI: 10.1016/j.trf.2014.04.005

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 vehicle automation levels and the severity of automation failures impact driver performance during critical driving situations. Motivated by the "out-of-the-loop" performance problem, where drivers become poor monitors of automated systems, the research hypothesized that semi-automated driving would be safer than highly automated driving when automation fails. It also examined whether the extent of deceleration failure (complete, severe, or moderate) influenced driver response. The experiment utilized a high-fidelity moving base driving simulator with 36 participants who had no prior experience with Adaptive Cruise Control (ACC). The study employed a mixed factorial design comparing two levels of automation: semi-automated (ACC only) and highly automated (ACC combined with automated steering, referred to as Traffic Jam Assist). Participants encountered three types of deceleration failures injected into traffic jam scenarios: Moderate (60% braking power), Severe (30% braking power), and Complete (0% braking power). Performance was measured using Point-of-No-Return (PoNR) events, Minimum Time-To-Collision (MTTC), Minimum Time Head-Way (MTHW), and Response Time (RT). Results indicated that driving performance degraded as the level of automation increased. Drivers in the highly automated condition experienced significantly more PoNR events and shorter MTTC values compared to those in the semi-automated condition. Specifically, the difference in PoNR events was statistically significant only during complete deceleration failures. Regarding the extent of failure, response times were significantly longer during moderate deceleration failures than during severe or complete failures. This suggests that drivers struggled more to detect and react to partial failures, likely because the system’s residual braking provided misleading cues about system functionality, whereas complete failures were more obvious. The findings support the hypothesis that higher levels of automation increase the risk of poor driver performance during automation failures, likely due to reduced situational awareness and complacency. The study concludes that while automation offers benefits, it shifts the driver’s role to a passive supervisor, making it difficult to detect subtle system malfunctions. These results highlight the need for improved human-machine interfaces and strategies to maintain driver engagement in highly automated vehicles to mitigate safety risks associated with automation failures.

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.

• automation complacency bias
• automation surprise
• situational awareness
• automation
• mode awareness
• takeover transitions

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, theory or model