AIoT-Based Fall Detection System for Enhanced Worker Safety in Confined Spaces

Accidents such as falls or abnormal movements in confined industrial spaces pose a constant threat to worker safety, making immediate detection and response essential. However, existing systems often generate false alarms, delaying critical actions and reducing their reliability. To address this issue, we developed a smart, wearable IoT solution that is practical, efficient, and highly reliable. Designed to be worn comfortably around the waist, the system uses two 3-axial accelerometers and microcontroller to monitor and analyze movement patterns in real time. When a fall or prolonged inactivity, potentially indicating unconsciousness, is detected, the system sends an instant alert to supervisors via the Pushbullet app, enabling quick response and reducing risks. Unlike traditional systems, our design eliminates the need for complex machine learning models, GPS modules, or additional hardware like buzzers, offering a lightweight and straightforward solution that integrates seamlessly into daily operations. Rigorous testing confirmed the device’s reliability and accuracy, proving its ability to deliver timely and actionable alerts. This innovative approach leverages IoT technology to create a cost-effective tool that enhances workplace safety, ensuring workers feel secure and supervisors can act swiftly in critical situations.