Citation

Nassereddine H, Santiago-Chaparro KR, Noyce DA. Transp. Res. Rec. 2022; 2676(11): 556-567.

Copyright

(Copyright © 2022, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.1177/03611981221094287

PMID

unavailable

Abstract

Surrogate safety measures (SSMs) have been used as an alternative to crash-based methods to study road safety. This study proposes a framework to estimate SSMs, such as post encroachment time (PET) and time to collision (TTC) values between right-turn-on-red (RTOR) and through vehicles, and demonstrates the feasibility of using vehicle trajectories obtained from existing radar-based vehicle detection systems to calculate such measures. The framework was implemented by the research team using a software-based approach. The framework presented consists of three stages: first, noise removal and classification of vehicles per lane; second, pairing RTOR vehicles and conflicting-through vehicles; and third calculation of PET and TTC values for each interaction using the radar-based vehicle trajectory dataset. Using video recordings, the interactions between right-turning vehicles and conflicting-through vehicles were documented and key vehicle position timestamps were logged. Logged timestamps enabled the calculation of PET. Video recordings acted as a tool to verify the results of PET calculations based on radar-based vehicle trajectory data. Therefore, the validity of the automated calculation of PET based on radar-based vehicle trajectory data was confirmed using the video recordings from the data collection site. The proposed framework demonstrates the potential to expand the capabilities of existing vehicle detection infrastructure to support proactive safety evaluations at a system level and increase the return on investment on existing vehicle detection technology. The proposed framework can be used in the development of safety models using existing vehicle detection infrastructure based on radar technology without the need for image processing techniques or specialized hardware.

Language: en