<![CDATA[Drowsiness and fatigue are the primary causes of traffic accidents among motorcycle riders. This study aims to design and test an Internet of Things (IoT)-based smart helmet for drowsiness detection to enhance riding safety. The system integrates an HC-SR04 ultrasonic sensor to detect yawning activity and an MPU6050 sensor to measure head tilt. Data from both sensors are processed by an ESP32 microcontroller using the Fuzzy Logic Mamdani method to classify drowsiness levels. The system is also equipped with a NEO-6M GPS module for location tracking and notification delivery via the Telegram application, along with physical warnings through a buzzer and vibration motor. Test results indicate that the reading accuracy of the HC-SR04 and MPU6050 sensors reaches 85.11% and 91.96%, respectively. The system successfully classifies rider conditions into normal, alert, and drowsy with an overall accuracy of 96.15%. This study proves that the integration of non-invasive sensors and fuzzy logic is highly effective as a preventive, affordable, and reliable early warning system to minimize the risk of microsleep-induced accidents among motorcycle riders.]]>
Copyrights © 2026
