Do adaptive cruise control and lane keeping systems make the longitudinal vehicle control safer? Insights into speeding and time gaps shorter than one second from a naturalistic driving study with SAE Level 2 automation

Varotto, Silvia F.; Mons, Celina; Hogema, Jeroen H.; Christoph, Michiel; van Nes, Nicole; Martens, Marieke H. · 2022 · Crossref

DOI: 10.1016/j.trc.2022.103756

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 whether Adaptive Cruise Control (ACC) and Lane Keeping Systems (LKS) enhance longitudinal vehicle safety by reducing speeding and time gaps shorter than one second. While previous research on early ACC systems indicated reduced short time gaps, most mathematical models assessing crash rates and congestion lack empirical grounding. This research addresses the gap in understanding how full-range SAE Level 2 automation influences driver behavior in naturalistic settings, specifically examining factors that trigger safety-critical behaviors like speeding and insufficient following distances. The researchers conducted a naturalistic driving experiment in the Netherlands involving ten participants driving leased BMW 5 Series and Tesla Model S vehicles equipped with full-range ACC and LKS. Data collection occurred between January 2018 and February 2019, comprising a one-month baseline period of manual driving and a two-month experimental period where participants used the automation systems as they deemed appropriate. Vehicle data, including speed, acceleration, and system states, were recorded at 10 Hz. The analysis focused on non-overlapping 10-second intervals on motorways with speed limits of 100 km/h or higher. Logistic regression models with random effects were employed to analyze the probability of speeding and short time gaps, controlling for variables such as traffic density, road characteristics, driver demographics, and system states. The findings reveal that, overall, drivers were less likely to speed and maintain time gaps shorter than one second when ACC and LKS were active compared to manual driving. However, specific conditions increased the likelihood of these behaviors. Drivers were more likely to speed in the subsequent 10-second interval if the current speed was near the limit, the upcoming speed limit was lower, or acceleration was high. Short time gaps were more probable when approaching a slower leader, when the current gap was already short, or during high acceleration. Crucially, drivers were significantly more likely to speed when overruling the ACC by pressing the gas pedal. Additionally, one vehicle brand exhibited a lower probability of short time gaps than the other, suggesting differences in system settings. Over time during the trip, the probability of speeding increased while the probability of short time gaps decreased after system activation. These results indicate that while SAE Level 2 automation generally promotes safer longitudinal control, driver interventions like overruling the system can reintroduce risk. The study highlights the importance of considering specific traffic and system states in safety assessments. The findings provide empirical data useful for designing automated vehicles that better prevent speeding and short time gaps, as well as for improving traffic simulations that evaluate the impact of ADAS on crash rates and congestion based on realistic on-road behavior.

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.

• following distance
• speed choice
• lane positioning

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, observational prevalence
• Methodological Resource: dataset resource