Towards Safer Driving: A Review of Real-Time Drowsiness and Hypoglycemia Detection Using Embedded Machine Learning and IoT-Based Alerts

Abstract

Traffic accidents often result in both human casualties and economic losses worldwide. Among the major causes of these accidents are driver drowsiness and hypoglycemia. In recent years, artificial intelligence technologies have significantly advanced and been applied across many domains, including road safety, particularly in detecting abnormal driver states such as drowsiness or low blood sugar. Computer vision techniques based on cameras have been used to monitor facial features such as blink rate, yawning, and head movement. Additionally, physiological sensors have been integrated into these systems using various biosensors. IoT technologies have also been employed to send alerts for remote driver monitoring, enhancing system reliability. This study aims to provide a comprehensive review of the technologies used to detect drowsiness and hypoglycemia in drivers and to highlight the progress made in this area. The goal is to emphasize the importance of research toward developing integrated systems that can simultaneously detect both conditions. The study concludes that integrating computer vision, biosensors, and machine learning algorithms can significantly contribute to the development of high-accuracy systems that improve road safety.