A domotics control tool based on MYO devices and neural networks

The automation of spaces has become a recurrent theme in current affairs due to the need to improve comfort, quality of life and facilitate work for the human being. Thus, this article proposes an intelligent system that allows controlling devices wirelessly in a domestic environment in a simple and safe way. Our system is based on the recognition of different gestures that user makes with his arm, using the bracelet MYO of the company Thalmic Labs. The bracelet consists of 8 electrodes, an accelerometer and a gyroscope. The implementation of the system is done through a wireless data collection classification module. The communication system is made up with ZigBee modules, which control the electrical and electronic devices in the home. In order to perform the recognizing and classification of electromyography (EMG) signals, an artificial neural network model based on supervised learning has used. This work specifies the procedure that we have followed to extract the characteristics of received signals, the training phase of learning system, and an explanation of used algorithms.