Repeated Route Naturalistic Driver Behavior Analysis Using Motion and Gaze Measurements

Abstract

Due to advancements in intelligent transportation and the emergence of automated vehicles, interest in analyzing driver behavior to improve commuters’ driving experiences has surged. Past studies have utilized driver gaze data to analyze behavior under various driving conditions using machine learning techniques. However, exploring driver behavior through multiple modalities can provide deeper insights. To this end, we conducted a naturalistic driver behavior study with ten participants, collecting vehicular data and driver gaze measurements using standard sensors. This dataset allows for an accurate assessment of driver behavior across different road types, traffic conditions, and congestion levels. Additionally, we investigated the influence of driving experience and time of day on behavior. Experienced drivers showed greater consistency across scenarios, while novices’ performance varied based on traffic intensity and route type.