Chen, Yuzhi (Southeast University), Xie, Yuanchang (University of Massachusetts Lowell), Wang, Chen (Southeast University), Xu, Sixuan (Southeast University), Wu, Lan (Nanjing Forestry University)

Temporal Dependency of Forward Collision Warning Effectiveness: A Functional Framework for Speed Profiles after Receiving Warnings

Scheduled for presentation during the Regular Session "Advanced Vehicle Safety Systems II" (ThAT8), Thursday, September 26, 2024, 11:10−11:30, Salon 16

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on October 8, 2024

Abstract

Forward Collision Warning (FCW) system is critical for improving road safety by alerting drivers to impending collisions. However, aggregated measures fail to capture the temporal dependency of FCW effectiveness after warnings are issued. This paper proposes a novel functional data analysis framework to investigate the temporal dependency of FCW effectiveness using real-world connected vehicle (CV) data. A functional representation procedure to observed CV data is designed to address irregular sampling intervals and the measurement errors from the real world. A nonparametric functional linear regression is introduced to quantify this temporal dependency. By modeling speed profiles after receiving warnings as continuous functions, our study suggests that FCW effectiveness is temporal-dependent exhibiting a trend of rapid increase followed by a gradual decline. The variation in FCW effects on drivers after warnings becomes more significant as time progresses. Moreover, the analysis of the functional coefficients suggests that average driver reaction time is approximately 1.3 seconds, and higher starting speeds significantly stimulate the driver intention to decelerate. These findings can support the optimization and validation of FCW systems, and also assist in calibrating driver response behavior to FCW warnings in simulated environments.