Design and Development of Real Time Bionic Hand Control Using EMG Signal

a large amount of study is carried out in the field of prosthetics to restore functionalities of lost organs. Bionic hand is one of those device which helps to replace the lost hand functionalities for the amputees. A lot of effort is put into to development of bionic hand which can able to mimic the action performed by normal hand. This paper provides an insight on development of real time control of bionic hand based on the collected Electromyography(EMG) signal from lower elbow amputee which able to perform hand opposition and re-position. This paper also provides an understanding of signal processing techniques required to classify the EMG signals for identifying required action to control bionic hand. The results indicate that root mean square and integrated absolute value for Feature extraction, k-Nearest Neighbor and Naive Bayesian Pattern Classification methods are chosen for feature classification EMG signals to control bionic hand. The developed algorithms are capable of producing accuracy up to 92-94%.