A Robust Human Fall Detection Wireless System
Nathaniel Varghese
Folsom, CA
Grade: 11
The project is an automatic and robust human fall detection system designed for elderly people. Today "pendants" exists for similar purpose, but these require the person to push a button when they fall. Often, when an elderly person falls, they are too incapacitated to do this, rendering such systems ineffective. This system uses an accelerometer as a sensor to automatically detect the fall. It uses a 4-stage fall detection algorithm running on an Arduino microcontroller that makes the fall detection robust with minimal false alerts. Finally, when a fall occurs, the system uses an XBee radio to send an alert signal wirelessly to a remote station to get help.