Feature: User fall detection

The user fall detection feature is designed to track current user three-dimensional position to determine user fall and additional help is required.

The methods which can be used to detect user fall are listed below.

Hardware sensor

Sensor	Advantages	Disadvantages	Details
phone accelerometer	scalability, mostly all the Android phones have accelerometer no specific hardware required	does not work if phone is not in you pocket/hand	override onSensorChanged() like in the example
smartwatch accelerometer	accurate	vendor specific, smartwatch has to have accelerometer proprietary API can be an impediment	vendor specific, request accelerometer data from smartwatch for example for garmin devices override onDataUpdate() and check RealTimeResult value

Software methods are the same for both of the hardware sensors.

Software method

Method	Details	Sources
simple algorithm	if vector sum of x,y,z <= 3 m/s than detect a free fall	Detection of falls using accelerometers and mobile phone technology
machine learning	collect accelerometer datasets with the different physical activity and build your own model	Detecting Falls Using Accelerometers by Adaptive Thresholds in Mobile Devices

Additional references:

• Automatic Fall Detection using Smartphone Acceleration Sensor

• Analysis of Android Device-Based Solutions for Fall Detection