Utilização dos sinais de eletroencefalograma e eletrodermal no aprendizado por reforço de uma interface cérebro-máquina

Abstract

The objective of this work is to construct an intelligent control system for a brain-machine interface using the Artificial Neural Networks paradigm. The built-in interface translates brain signals to move a cursor on a digital screen. The control system uses a feedback signal from the user to adjust the cursor movement in a personalized way according to the signals that the user sends. With the use of artificial neural networks we have been able to reduce the training time from up to 45 days, in traditional control systems, to less than 10 minutes. The project aims to facilitate the accessibility of individuals who have limitations of their physical and motor capacity, whether temporary or permanent. The construction of the limbic signal translator system in digital responses allows the development of a range of new applications to increase the autonomy of people with motor limitations. As a continuation of this work, many automation applications may be developed, for home or hospital use. By using low cost platforms, it has great potential for production and distribution. . In some researches, the Brain Computer Interface (BCI) system has proven to be a promising tool in applications that assist people with severe motor limitations and in health care devices, remote or otherwise.