Driver Responses to Cyclists Crossing from Near and Far Side in Driving Simulator Experiments

Zhao, Yuqing; MIZUNO, Koji; Aoshima, Yuhei; Kanianthra, Joseph N.; Inagami, Makoto · 2022 · OpenAlex-citations

DOI: 10.20485/jsaeijae.13.2_89

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 driver responses to cyclists intruding into a vehicle’s path at intersections, aiming to inform the development of active safety systems like autonomous emergency braking (AEB). Motivated by the high prevalence of car-to-cyclist collisions in Japan, particularly at intersections, the research addresses the gap in understanding how drivers react to cyclists emerging from both the near side (same side as the driver) and the far side (opposite side), which often requires peripheral vision. The researchers conducted experiments using an immersive driving simulator with 31 licensed participants. Three scenarios were tested: two near-side approaches with time-to-collision (TTC) values of 1.53 s and 0.84 s, and one far-side approach with a TTC of 2.07 s. Eye-tracking technology measured gaze direction and notice time, while vehicle kinematics recorded braking reaction time (BRT), swerving, and deceleration. Statistical analyses, including decision trees and logistic regression, were employed to determine the factors influencing collision occurrence and avoidance. The results indicate that braking was the primary effective response, whereas swerving alone failed to avoid collisions in all instances. A significant difference was observed in BRT based on the cyclist's approach direction: the mean BRT was 0.65 s for near-side scenarios and 1.35 s for the far-side scenario. This delay in the far-side scenario was attributed to longer notice times, which correlated with the angle between the driver’s initial gaze and the cyclist’s position. Drivers often gazed toward their future path or the left side, delaying detection of cyclists emerging from the right (far side). The study established that collisions occurred when the sum of BRT and braking deceleration time exceeded the TTC. A decision tree model identified TTC and BRT as the most critical predictors of collision, with an 87.1% prediction accuracy. Specifically, if TTC was less than 2.19 s, collision avoidance probability dropped significantly if BRT exceeded 0.558 s. The significance of this work lies in its formulation of a collision occurrence equation based on TTC, BRT, and vehicle velocity, which can help define "collision areas" for safety system design. The findings highlight that current pedestrian-focused safety systems may be insufficient for cyclists due to their higher speeds and the drivers' delayed recognition of far-side intrusions. By quantifying the impact of gaze direction on reaction times, the study provides critical data for improving Advanced Driver Assistance Systems (ADAS) to better detect and respond to cyclists in complex intersection environments.

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 2026-06-26; verification: verified.

Topics

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

• looked but failed to see
• time to collision looming
• driver vru interaction
• hazard perception

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