Friendly Robotic Arms Which Controlled by Electroencephalography

Abstract

Nowadays, many rehabilitation robots are based on traditional human-computer interaction method such as by voice and by push button. However, there are a large amount of disabilities and elders who are lack of language competence and physical ability for many reasons. Thus, a robotic arm which directly controlled by one’s own brain instead of others like nurses may possibly be more effective to help them to grab things and prevent them from falling during rehabilitation, and electroencephalography (EEG) controlled robot will be required. This abstract-based poster shows the working process of the specific kind of mechanical arms which based on brain-computer interface (BCI) and controlled by EEG. When people concentrate their minds to reach a certain purpose, a relatively constant specific signal will appears in the form of EEG. This signal may be not as constant and stable as we expected at the beginning of the trial, therefore several practices may be needed in order to enhance the stability of this specific signal, which will be collected by a sensor which is connected to the personal computer through the Application Programming Interfaces (API) at the real time. After the analog signals are collected by the sensor, they will be transferred into a Robot Operating System (ROS), which is an open source software platform which works as a transition station that convert analog signals to digital signals which can directly apply to the robotic arm. The robotic arm, which is the terminal actuator, has 7 degrees of freedom and can mimic the movement of human arms by receiving the digital signals as the command.