Enhancing Intersection Traffic Safety Utilizing V2I Communications: Design and Evaluation of Machine Learning Based Framework

Shahriar, Mohammad Sajid; Kale, Arati Kantu; Chang, Kyunghi · 2023 · DOAJ

DOI: 10.1109/ACCESS.2023.3319382

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the critical challenge of improving intersection traffic safety, a major source of motor vehicle fatalities and injuries. While Cooperative Intelligent Transport Systems (CITS) utilizing Vehicle-to-Everything (V2X) communications promise significant reductions in accidents, their effectiveness is currently limited by low penetration rates of V2X-enabled vehicles and unresolved questions regarding system performance under real-world constraints. The authors propose the Intersection Traffic Safety Framework (ITSF), a safety-oriented system designed to mitigate collisions at road intersections by leveraging Vehicle-to-Infrastructure (V2I) communications and machine learning. The framework aims to distinguish between vehicles posing a collision risk and those that do not, thereby managing computational complexity and reducing unnecessary alert transmissions. The study focuses on collision scenarios involving reckless driving behaviors, particularly during yellow traffic light phases where drivers may accelerate to cross intersections before the light turns red. The ITSF incorporates two key components: a risk assessment algorithm and an alert dissemination mechanism. The risk assessment utilizes vehicle-based measures such as position, heading, Time-to-Collision (TTC), and environmental factors like traffic phase and remaining time for phase changes to model the relative state of vehicles. A machine learning model is employed to classify vehicles into "subject" (approaching the yellow light) and "object" (waiting at the red light) categories, identifying high-risk interactions. The framework evaluates performance by analyzing critical factors including the penetration rate of V2X-enabled vehicles, end-to-end communication latency, beaconing range, and driver responsiveness to safety alerts. The results demonstrate that the proposed ITSF is highly effective in ensuring road user safety when the penetration rate of V2X-enabled vehicles reaches 60% or higher. Under these conditions, the framework achieves a significant reduction in collisions and maintains nearly 98% accuracy in classifying risky vehicles. The study further reveals that the algorithm remains robust even in scenarios where some drivers neglect or ignore safety warnings, showcasing its ability to minimize collision rates despite human error. The evaluation highlights that while higher penetration rates enhance system efficacy, the framework provides substantial safety benefits even before widespread adoption, addressing a gap in previous research that often assumed near-universal V2X deployment. The significance of this work lies in its comprehensive approach to intersection safety, bridging the gap between theoretical collision avoidance algorithms and practical implementation challenges. By integrating machine learning for precise risk classification and evaluating the impact of communication latency and partial vehicle penetration, the ITSF offers a scalable solution for current traffic environments. The findings suggest that V2I-based safety systems can be effectively deployed in the near future without waiting for the projected 15-year timeline for high V2X penetration rates. This contributes to the field by providing a validated framework that accounts for dynamic vehicle interactions and human factors, thereby enhancing the reliability and practicality of intelligent transport systems in reducing intersection-related fatalities.

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.

• v2x connected vehicle