Hazard anticipation is a critical skill for crash avoidance, requiring drivers to detect, predict, and respond to potential roadway threats. As Advanced Driver Assistance Systems become more common, drivers increasingly shift from active vehicle control to supervisory roles, which may change visual scanning patterns, situational awareness, and response strategies. These changes are not yet fully understood and may have important safety implications.
This project investigates how automation affects hazard anticipation by examining driver gaze behavior, responses to traditional roadway hazards, and reactions to system-related cues in ADAS-equipped vehicles. The research integrates secondary data analysis, development of a hazard anticipation taxonomy, and experimental testing using a high-fidelity driving simulator. Results will be used to identify automation-related changes in hazard anticipation and to develop targeted training approaches aimed at maintaining effective driver oversight. Findings from this work will support safer human interaction with automated vehicle technologies as their use continues to expand.