Studying the Simultaneous Effect of Autonomous Vehicles and Distracted Driving on Safety at Unsignalized Intersections

Khashayarfard, Mohammad; Nassiri, Habibollah · 2021 · Crossref

DOI: 10.1155/2021/6677010

archive: archived pipeline: cataloged

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

Summary

**Research Question and Motivation** This study investigates the simultaneous impact of Autonomous Vehicles (AVs) and driver distraction on traffic safety at unsignalized intersections. While previous research often modeled AV behavior in isolation or assumed zero probability of human driver distraction, this paper addresses the gap by integrating calibrated driver distraction parameters into microsimulation models. The motivation stems from the fact that human error, particularly distraction from mobile phone use, is a leading cause of accidents, and understanding how AVs interact with distracted human drivers is critical for assessing the safety benefits of automation in mixed-traffic environments. **Methods and Experimental Design** The researchers utilized PTV Vissim software to simulate two unsignalized intersections in Tehran, Iran: Vesal Shirazi-Bozorgmehr and Sattarkhan-Niroo. Traffic volumes and speed distributions were calibrated using field data collected during peak hours. To accurately model human-driven vehicles, the study incorporated driver distraction parameters derived from a driving simulator experiment involving 20 participants. These parameters included the probability of distraction, the duration of distraction (ranging from 0.64 to 7.34 seconds), and lane deviation angles. For AVs, the study compared two European projects: the "Coexist" project (featuring four driving patterns from cautious to all-knowing) and the "UK Autodrive" project. Five scenarios were simulated based on AV market penetration rates: 0%, 25%, 50%, 75%, and 100%. Safety was assessed using the Surrogate Safety Assessment Model (SSAM), specifically employing Time to Collision (TTC) and Deceleration Rate to Avoid Crashes (DRAC) indicators to quantify potential conflicts. **Findings** The simulation results indicate that the presence of AVs significantly reduces the potential for accidents. Specifically, achieving a 100% penetration rate of AVs reduced the number of potential conflicts by up to 93% compared to the baseline scenario with only human-driven cars. The study found that even at lower penetration rates, there was a significant reduction in accidents at unsignalized intersections, contrasting with some findings at signalized intersections where low AV percentages might increase risk due to dilemma zones. The comparison between the two AV projects revealed differences in their interaction with conventional vehicles, with the "All-Knowing" AVs from the Coexist project demonstrating the ability to maintain smaller headways and higher network capacity compared to more conservative AV types. **Significance** This research provides evidence that the safety benefits of AVs are maximized when they are integrated into networks where human driver behaviors, such as distraction, are explicitly modeled. It highlights that unsignalized intersections, which are often hazardous due to driver noncompliance, see substantial safety improvements with AV adoption. The findings suggest that transportation planners should consider the specific capabilities of different AV generations when forecasting safety outcomes, as the interaction between AVs and distracted human drivers is a key determinant of overall network safety.

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 qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.

Topics

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

• situational awareness