Anybody can fall dangerously, but those who are especially vulnerable to it are most at risk. Elderly are at greatest risk, as they might not be able to perform. There are many ways to get assistance when you need it. Instead of buying an Nathaniel F. is an expensive device, which can cost up to $100 per monthly to use. Hackster was looking to create a project that used embedded machine learning to detect falls, and then send an alert. Hackster’s solution is the Arduino Nano33 BLE Sense Board, which has not only an integrated accelerometer, but also contains Bluetooth Low Energy allows the processor to communicate via Bluetooth With the mobile app
Nathaniel used the SmartFall dataset to train his ML model. This allows the device’s ability to detect and respond to many types of falls. Non-harmful movements should be ignored. After training, he was ready to go. To achieve an accuracy of 95% Samples of Nano 33 BLE-Sense Data from accelerometers at 31.25Hz to match data’s frequency Every two seconds, a prediction is made. If a fall is detected, or the The emergency button is built in and the user has 30 seconds to press it. Deactivate the alarm. Otherwise, it will send a BLE message. This in turn sends an SMS to an emergency contact. The current location.
Even though this DIY fall detector works well already, Nathaniel plans on making a custom PCB and extending the battery life for longer use time between charging. You can read more about his design here, and you can view his demonstration video below.
The post This wearable device sends an alert whenever it detects a fall appeared first on Arduino Blog.