Safety Impacts of Automated Vehicles in Mixed Traffic

Automated Vehicles can enhance the current state of transportation by reducing human errors and improving safety and mobility. This paper presents a study that predicts the impact of automated vehicles on the safety performance at intersections. As a motivation for the study, real-life AV-involved crashes in California are analyzed. Thirty-one crashes were reported among which 22 automated vehicles were operating in the autonomous mode at the time of the collision. The crashes were predominantly rear-ends at intersections. To anticipate future safety impacts of mixed traffic at intersections, a simulation framework is defined that considers travelers who are not driving AVs and those who have some level of automation, relieving them of car-following and steering tasks. Different market penetration scenarios are simulated using SUMO (Simulation of Urban Mobility) software. The Wiedemann car-following model is suitable for the study as it allows simulating both traffic flow and traffic crashes. However, the default Wiedemann model in SUMO does not reflect the real-world relationship between vehicles accelerations and speed. Therefore, the Wiedemann model was modified and tuned using real-world Basic Safety Message (BSM) data. It was then used to characterize the behavior of human-driven vehicles and AVs. The case study that simulates the mixed environment of human-driven, level 3, and level 5 AVs indicates that with decrease in human-driven vehicles, the crash rate will decline. However, when it comes to the mixed environment of human-driven and level 5 vehicles, the safety benefits will be substantial at about 40% market penetration levels.

• Supplemental Notes:
  • This paper was sponsored by TRB committee AHB45 Standing Committee on Traffic Flow Theory and Characteristics.
• Authors:
  • Arvin, Ramin
  • Kamrani, Mohsen
  • Khattak, Asad J
  • Rios-Torres, Jackeline
• Conference:
  • Transportation Research Board 97th Annual Meeting
  • Location: Washington DC, United States
  • Date: 2018-1-7 to 2018-1-11
• Date: 2018

Language

• English

Media Info

• Media Type: Digital/other
• Features: Figures; References; Tables;
• Pagination: 5p

Subject/Index Terms

• TRT Terms: Automation; Highway safety; Intelligent vehicles; Vehicle mix
• Geographic Terms: California
• Subject Areas: Highways; Safety and Human Factors;

Filing Info

• Accession Number: 01657930
• Record Type: Publication
• Report/Paper Numbers: 18-00088
• Files: TRIS, TRB, ATRI
• Created Date: Jan 25 2018 9:34AM