Distracted driving emerges as a global threat, significantly contributing to the alarming toll of 1.3 million annual traffic fatalities. This paper presents an innovative solution employing a Unity-based driving simulator with biometric features to tackle distracted driving across educational and technological domains. The simulator uses the popular Mediapipe Solutions library and uncomplicated camera setups to capture pivotal biometric parameters: head rotation, gaze direction, and eyelid opening. The fusion of these parameters creates an immersive user experience, enabling self-assessment of distraction levels within simulated nighttime scenarios. The simulator incorporates alerts for incorrect gaze direction or signs of drowsiness, employing an acoustic signal. Furthermore, the simulator activates car headlights upon the driver’s proximity to the dashboard, indicating compromised visibility. The proposed solution’s efficacy is confirmed through experiments conducted under diverse conditions, including scenarios with sunglasses, eyeglasses, and low luminosity. With minimal hardware and software requirements, the simulator emerges as a valuable educational tool for drivers, holding potential for integration into assisted driving systems. The results highlight its significant contribution to road safety, effectively addressing the pervasive issue of distracted driving through a comprehensive and accessible framework.
A Biometric-Based Adaptive Simulator for Driving Education
Bisogni C.;Pero C.
2024
Abstract
Distracted driving emerges as a global threat, significantly contributing to the alarming toll of 1.3 million annual traffic fatalities. This paper presents an innovative solution employing a Unity-based driving simulator with biometric features to tackle distracted driving across educational and technological domains. The simulator uses the popular Mediapipe Solutions library and uncomplicated camera setups to capture pivotal biometric parameters: head rotation, gaze direction, and eyelid opening. The fusion of these parameters creates an immersive user experience, enabling self-assessment of distraction levels within simulated nighttime scenarios. The simulator incorporates alerts for incorrect gaze direction or signs of drowsiness, employing an acoustic signal. Furthermore, the simulator activates car headlights upon the driver’s proximity to the dashboard, indicating compromised visibility. The proposed solution’s efficacy is confirmed through experiments conducted under diverse conditions, including scenarios with sunglasses, eyeglasses, and low luminosity. With minimal hardware and software requirements, the simulator emerges as a valuable educational tool for drivers, holding potential for integration into assisted driving systems. The results highlight its significant contribution to road safety, effectively addressing the pervasive issue of distracted driving through a comprehensive and accessible framework.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
