ML with Sensors

In this repo, I will teach you how to get started Machine Learning with your own sensors. Here we are sending data from a sensors to a server wirelessly. The server is processing the data and extrating meaning out of that data using Machine Learning.
This is one of the sample board created by me with Arduino Pro Mini, MPU9020 and ESP8266. But in this, we will use Arduino Nano because it does not required external programmer unlike Arduino Pro Mini and MPU6050 instead of MPU9020.

Step 0 >> Componets you will need.

Arduino (here Arduino Nano) , A Sensor (here MPU6050) , ESP8266-01

Step 1 >> Making your ESP ready to send the data.

Sending sensor data to UDP Server.

• Upload a Bare Minimum Code for your Board.

• Choose your board that you will be using to program your ESP. e.g Arduino Nano.

• Download ESP8266 library & Board and add to the Arduino IDE.

  • Open the preferences window from the Arduino IDE. Go to File > Preferences
  • Enter http://arduino.esp8266.com/stable/package_esp8266com_index.json into Additional Board Manager URLs field and click the “OK” button
  • Open boards manager. Go to Tools > Board > Boards Manager…
  • Scroll down, select the ESP8266 board menu and install “ESP8266 platform”

• Select Board as Generic ESP8266 and Upload the esp.ino file to the ESP. Make these changes to the esp.ino file before uploading

• Steps for making changes in esp.ino file.

  • Make a hotspot using your phone.
  • Set SSID and Password in your phone for hotspot.
  • Connect your laptop to same hotspot and find your IP Address. It will look something like this- "192.168.43.101". In Linux you can do this by pressing ifconfig in Terminal.
  • Finally, replace your SSID, Password, UDP address (same as IP address) and UDP Port (your choice four digit number greater than 1024, like 6969)
  • Save the changes and now your are ready to upload this file to the ESP Board.

  Connections for uplaoding esp.ino in ESP
  Tx - Tx
  Rx - Rx
  VCC- 5V
  GND - 0V
  GPIO0 -0V
  CH_PD - 5V
  Arduino Reset- Arduino GND

• Once uploaded,disconnect your Arduino Board.

• Remove Arduino Reset from Ground.

Step 2 >> Making your sensor ready.

Connections
SCL - A5
SDA - A4
INT - D2
VCC- 5V
GND - 0V

• Copy I2Cdev and MPU6050 to your Arduino's libraries folder and restart the IDE.
• Open the Arduino IDE and got to Examples>MPU6050>MPU6050_DMP6
• Upload this file to Arduino Nano and check on serial monitor if MPU is giving the data.
• Connect the ESP8266 according to following connections.
  Connections
  Tx - Rx
  Rx - Tx
  VCC- 5V
  GND - 0V
  GPIO0 -open
  GPIO1 - open
  CH_PD - 5V

Step 3 >> Send data to UDP Server.

• Mak

Step 4 >> Save the incoming data. (server side)

Step 5 >> View Data.

Step 6 >> Filtering and Trimming the Data.

Step 7 >> Feature Extraction and Labelling.

Step 8 >> Training your Classifier and exporting the Model.

Step 9 >> Realtime Predictions.