AANN

Artificial Neural Network implemented and run using an Arduino Nano.

Included is an implementation of a perceptron in nano/ann.ino which uses a switch button as input signal and a LED light as a correct classification signal. The algorithm dynamically updates it's weights using online gradient descent with a configurable learning rate.

There is also a collection of various scripts related to the Arduino Nano and Uno boards in their respective folders.

Setup

1. You will need to install the Arduino IDE or Web Editor to interface with the boards.

2. Assemble the circuit following the diagrams named layout.png or link at the top of the .ino script.

3. Upload the code contained in this sketch on to your board using the IDE or Web Interface.

This will execute the code on the microcontroller.

Materials

• Arduino Uno board (standard)

• Arduino Nano board (AT328P flavour)

• Breadboard

• Jumper cables

• Charging cables

  • USB 2.0 to Micro USB (Nano)

  • Standard A-B USB cable (Uno)