Simulated prosthetic hand actuated by Electromyography

[Eric]

The muscular activity in the arm is detected through electrodes using Electromyography. A machine learning algorithm is classifying (for the moment) 6 states triggering 5 actions for the prosthetic

• _ open Palm Grip
  _ active Index Grip
  _ power Grip
  _ mouse Grip
  _ trigger

Trigger states is triggering a sub action when the current action has a sub action associated (e.g. when in mouse grip, trigger state generates a motion of the index to click the left button of the mouse)

The controller was developed with a Tiva C Launchap ARM Cortex-M4F on Keil uVision MDK based on DSP and Machine Learning techniques.

https://youtu.be/q5snFAOdzsI