Falls are a major cause of injury and disability among the elderly people which compromises their independence and overall quality of life. This paper presents an innovative Elderly Fall Detection and Assistance System by utilizing IoT technology for real-time monitoring and immediate response to fall events. The system is designed with a central processing unit and inertial sensors to monitor movement and orientation for fall detection. MQTT is being used for the wireless communication to deliver the alert notifications along with a user-friendly interface for smooth interaction. The fall detection algorithm identifies the sudden changes of the motion and the posture of the device which triggers the real-time alerts to caregivers including a visual indicator. Extensive testing has been done by simulating falls and routine activities which has exhibited the system’s reliability, minimized false positives, and ensured timely notifications. Moreover, this detection system can offer a cost-effective solution compared to many existing options on the market, with potential for future advancements in enhancing accuracy and adaptability.

Authors:
Asma Ahmed, McMaster University, Canada
Shuning Wang, McMaster University, Canada
Mengmei Xu, McMaster University, Canada
Marjan Alavi, McMaster University, Canada

About the Presenter(s)
Asma Ahmed is a graduate student at McMaster University. Her primary interest is in AI, ML/DL, IoT and data analysis.

See this presentation on the full schedule – On Demand Schedule

Bookmark