EMG decoding based on Arduino electrodes

Introduction

The idea is to find a good setup to retrieve and measure the quality of EMG signals using Myoware sensors for further applications, for instance to control a robotic hand prosthesis based on forearm muscle contractions. The obtained results show that these sensors could be as good as the Noraxon sensors, with SNR value up to 6.8.

Organisation

This project is organized as follows :

• the repository src that includes several files trying to implement some signal processings and collection of the data to send to a stream outlet.
• the file Bachelor project report - Changling Li.pdf which is the report of this project that provides a full explanation of what has been done.
• the file collectsendEMG.py collects the EMG datasamples from the Arduino to send them into a stream outlet using the liblsl library. From there, any computer connected to the stream can retrieve the datasamples.
• the file send_bytes.py.ino collects the EMG datasamples from the Myoware sensors via Arduino to send them to the computer.

How to use this project

After placing the Myoware sensors on your forearm, run the file send_bytes.py.ino in the Arduino IDE. Then, run the file collectsendEMG.py in the terminal. Once you are finished with the data collection, type Ctrl + C to stop the program.

Libraries

• serial
• liblsl