Evaluation of an Infrastructure-Based Warning System: A Case Study on Roundabout Driving Behaviors

Smart intersections have the potential to improve road safety with sensing, communication, and edge computing technologies. Perception sensors installed at a smart intersection can monitor the traffic environment in real-time and send infrastructure-based warnings to nearby travelers through vehicle-to-everything (V2X) communication. This study investigated how infrastructure-based warnings can influence driving behaviors and improve roundabout safety through a driving simulator experiment. A co-simulation platform integrating Simulation of Urban Mobility (SUMO) and Webots was developed to serve as the driving simulator. A real-world roundabout in Ann Arbor, Michigan was built in the co-simulation platform as the study area, and merging scenarios were investigated. 36 participants were recruited and asked to navigate the roundabout under three aggressiveness levels (low, medium, and high) and three collision-warning designs (no-warning, 1-second-in-advance warnings, and 2-second-in-advance warnings). Experiment results indicate that advanced warnings can significantly enhance safety by minimizing potential risks compared to scenarios without warnings. Earlier warnings enable smoother driver responses and reduce abrupt decelerations. In addition, a personalized intent prediction model was developed to predict drivers’ stop-or-go decisions when the warning was displayed. Among all tested machine learning models, the XGBoost model achieves the highest overall prediction accuracy.