Roadsense: An Android-Based Drowsiness Detection System for Preventive Safety Measures

Abstract

Drowsy driving significantly contributes to global road accidents, yet advanced detection technologies remain exclusive to high-end vehicles, limiting their accessibility in developing countries like the Philippines. To bridge this gap, the researchers developed this project, an Android-based application designed to detect driver drowsiness in real-time and provide auditory alerts. The study aimed to identify user requirements, develop the system using the Scrum Framework, and evaluate its usability, performance, and detection accuracy. The descriptive, developmental, and experimental research designs were used in this study, involving 20 respondents for requirements elicitation and 10 participants for performance testing. The researchers used (1) descriptive research to gather user requirements, (2) developmental research, specifically the Scrum Framework to ensure iterative software development, and (3) experimental research to test the system's performance under different lighting conditions. The findings showed that the system is highly usable, providing an application for drivers seeking preventive safety measures. Furthermore, it was determined that the drowsiness detection feature performs well in well-lit environments. However, its accuracy decreases in dim-lit conditions due to the limitations of smartphone camera sensors and environmental conditions. It was concluded that the application serves as a functional solution for alerting users of their drowsiness by combining AI-driven monitoring with navigation services. This project demonstrates the potential of mobile technology to improve driver awareness and road safety management.