“Introducing GetFit, your ultimate health and workout partner. GetFit is an easy-to-use, teachable fitness tracker with the capability of detecting an endless variety of exercises. Powered by Arduino Nano 33 BLE Sense and Edge Impulse, it is a completely open-source project.
- Can Count unlimited exercises
- Fully open-source
- Calorie Burn Estimation on a daily and weekly basis
For doing this project we are using, Arduino nano 33 BLE sense. It’s a 3.3V AI-enabled board in the smallest available form factor. It comes with a series of embedded sensors.
- LSM9DS1 (9 axis IMU)
- LPS22HB (Barometer and temperature sensor)
- HTS221 (relative humidity sensor)
- APDS-9960 (Digital proximity, Ambient light, RGB, and Gesture Sensor)
- MP34DT05 (Digital Microphone)
Here we are utilizing the 3 acceleration channels of the LSM9DS1 sensor for counting the activities.
It is a very hard task to recognize the activities using rule-based programming, as people don’t perform activities, in the same way, every time. But machine learning can handle these variations with ease. To create a machine learning model, you would traditionally use a framework like TensorFlow or Scikit-learn on top of a high-level language like Python. There are a lot of benefits to still learning some of these frameworks, but here we’re going to use a tool called Edge Impulse as it just makes model training much easier. Fortunately, the Arduino Nano 33 BLE Sense is fully supported by Edge Impulse.”