By Jeremy Ellis
Started July 7th, 2022
Use at your own Risk!

These are examples using the Nano 33 Ble Sense that are not the standard: Acceleration, Sound and Vision that students would use combined with EdgeImpulse.com University or TinyML4D

1. a01-multi-sensor-nano33BleSense.ino Just a fun plotter program showing all the Senses on the Nano33BleSense, for which each sense is adjusted slightly so it has it's own row on the serial plotter.

    barometricPressure = BARO.readPressure()*-5;
    temperature = HTS.readTemperature()*-10;
    humidity = HTS.readHumidity()*-5;
    
    accelerometerX = 9.8 *accelerometerX + 100;   // acceleration to Force F = gA = 9.8 *a
    accelerometerY = 9.8 *accelerometerY + 200;  
    accelerometerZ = 9.8 *accelerometerZ + 300;  
    

2. a02a-3axis-data-forwarder.ino
   a02b-3axis-acceleration-as-raw-sensors.ino
   Just getting the EdgeImpulse accelerometer as a raw sensor data forwarder and Classification working. An edgeimpulse model using the raw data would have to be made. This code using the accerlometer but is a bridge for using any raw sensor.

3. a03a-color-data-forwarder.ino
   a03b-color-as-raw-sensor.ino
   Same as the accelerometer above but this time using raw color RGB data. Makes the transition to any 3 raw sensors.

4. a04a-one-sensor-proximity-data-forwarder.ino
   a04b-one-sensor-proximity-classify.ino
   a04c-single-buffer-proximity-classify-continuous.ino
   a04d-double-buffer-proximity-continuous.ino
   Now using a single raw senosr this time the proximity sensor with an edgeimpulse regression raw model, measaure punch speeds (no units). Then the classification gets more complex. a04b is regular, a04c is continuous using a single buffer and a04d uses a double buffer.

5. Latest work July 28, 2022 a05a-10-motion-proximity-multi-sensor-data-forwarder.ino Now prepping for the data forwarder to take 10 samples of raw data. this time proximity, 3 acceleration, 3 Gyrocope, 3 magnetic. 10 total.
   a05b-10-motion-proximity-multi-sensor-data-forwarder.ino Presently in DRAFT mode, but this is the basic classification after an edgeimpulse.com model has been made with the above data forwarder. This is just a simple test to see if it works. Nice thing with this is the code is reasonably easy to work with.
   I am much more interested in the continuous abilitiy which I will work on next.

All that needs to be done is more examples with more working code.

z01-double-buffer-testing.ino This is just a testing of how to do a double buffer. I put a counter into the code so I could see the numbers going up one at a time. Also highlighted the issue of the last few values for a main buffer that was not fully divisable by the smaller buffer.

Which sensors are available other than motion, sound and vision? humidity, temperature, barometric, gesture, proximity, light color and light intensity sensor

Good video

https://www.youtube.com/watch?v=7ZncQCU9H6w

One of my multi-sensor videos for the Nano33ble sense

https://www.youtube.com/watch?v=gNRIQ5clpkY

Libraries

LSM9DS1 - IMU (Accelerometer, gyroscope, magnetometer)
LPS22HB - Barometer Adafruit PDM - Microphone