Modelling overtaking strategy and lateral distance in car-to-cyclist overtaking on rural roads: A driving simulator experiment

Farah, Haneen; Piccinini, Giulio Bianchi; Itoh, Makoto; Dozza, Marco · 2019 · Crossref

DOI: 10.1016/j.trf.2019.04.026

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 addresses the safety risks associated with car-to-cyclist overtaking on rural roads, where interactions often result in high-severity crashes. The research aims to develop predictive models for two key aspects of driver behavior: the choice of overtaking strategy (flying vs. accelerative) in the presence of oncoming traffic, and the lateral comfort distance maintained during the maneuver. Understanding these behaviors is critical for designing active safety systems, such as forward collision warning and automated emergency braking, and for refining the behavioral algorithms of automated vehicles. The researchers conducted a driving simulator experiment involving 37 Japanese drivers who performed seven overtaking maneuvers each on a two-lane rural road with oncoming traffic. The study distinguished between "flying" overtaking, where drivers maintain speed, and "accelerative" overtaking, where drivers slow down significantly before passing. Data were analyzed using binary logistic regression models with mixed effects to predict the overtaking strategy, and linear mixed models to predict lateral comfort distance. The analysis incorporated variables such as longitudinal distances to the cyclist and oncoming vehicle, relative speeds, and driver characteristics derived from questionnaires on sensation seeking and ordinary violations. The results identified 259 overtaking maneuvers, with 168 classified as flying and 91 as accelerative. Driving speed was found to be a significant factor influencing the choice of strategy. The predictive models for overtaking strategy demonstrated strong performance, achieving an accuracy of 85–90%. In contrast, the models predicting lateral comfort distance were less accurate, with root mean square errors ranging from 0.56 to 0.62. The lateral distance was primarily influenced by the longitudinal distances to the oncoming vehicle and the cyclist, as well as the presence of oncoming traffic. Additionally, driver personality traits played a role: sensation seeking correlated with lateral distances in flying maneuvers, while ordinary violations affected distances in accelerative maneuvers. The study concludes that while predictive models for overtaking strategy are robust enough to support the development of active safety systems and automated driving policies, models for lateral distance require further refinement. The findings highlight that drivers trade off lateral space based on the proximity of oncoming traffic and their own risk preferences. These insights provide a foundation for creating automated vehicle behaviors that mimic human decision-making processes, potentially increasing driver trust and improving cyclist safety by ensuring appropriate lateral clearances are maintained during overtaking.

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 2026-06-10; verification: verified.

Topics

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

• following distance

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).

• Theoretical Contribution: computational model