Brain Machine Interface for wrist movement using Robotic Arm

Brain Machine Interface (BMI) has made it possible for the disabled people to communicate with the external machine using their own senses. In the field of BMI, the invasive techniques have been widely used. This paper deals with the study of features of Electroencephalography (EEG), a non invasive technique that has been used for classifying two classes of movements, namely Extension and Flexion. Classification of movements is done on the basis of energy, entropy, skewness, kurtosis and their various combinations. The maximum accuracy of 91.93% has been obtained using discrete cosine transformation of energy and entropy. Finally the detected wrist movement is implemented on a mechanical Robotic Arm using ARDUINO UNO and MATLAB.