Development of a Driver Drowsiness Detection System Using Infrared, Tilt and Heart Rate Sensors

Abstract

Driver drowsiness is a major contributor to road accidents, making early detection systems essential for enhancing vehicle safety. This paper presents the design and evaluation of a wearable drowsiness detection prototype integrating an infrared (IR) sensor for eye blink monitoring, an MPU6050 accelerometer for head tilt detection, and a MAX30100 sensor for heart rate and oxygen saturation measurement. Laboratory validation was first carried out using servo motors and oscilloscope readings to simulate eyelid movement and head tilts. The system was then tested in-vehicle under real driving conditions, where real-time monitoring of eye status, head posture, and heart rate was displayed on an LCD with buzzer and LED alerts. Results demonstrated reliable detection of prolonged eye closure, abnormal head tilts, and reduced heart rate, with an estimated overall accuracy of approximately 85%. These findings confirm that combining behavioral and physiological indicators enhances detection robustness compared to single-parameter systems. The prototype demonstrates a low-cost, portable, and non-intrusive solution for drowsiness detection, making it suitable for vehicles lacking advanced driver-assistance systems. Furthermore, the use of easily available components allows for wide applicability and integration into various vehicle types. Future research will focus on refining adaptive thresholds, conducting long-term driving trials, and applying machine learning techniques to further improve detection reliability.