Design and implementation of a laboratory training device to simulate partial finger movement using an EMG intelligent controller

The human hand is formed with one thumb and four fingers; a single finger can thus be seen as a starting point for the entire design process of creating a hand based on replicating the design of the finger. This paper deals with the process of designing and implementing a prosthetic finger for a partial hand/finger amputee for the simulation and training of students in prosthetics and orthotic engineering laboratories. In order to help such students understand how to compensate for the finger or fingers of the hand in terms of restoring functional performance, this work presents the manufacture of finger with joints that mimic real human fingers using 3D printing technology. An Arduino with servomotors was used to facilitate finger part movements based on the use of an intelligent controller. A myo-arm band sensor was used to extract electromyography (EMG) signals for finger gestures, which were then classified using an artificial neural network (ANN). In this research, the implemented training device was focused on the amputated or partially amputated finger, as the loss of a part of the finger or the finger as a whole reduces the functional performance of the hand, and thus this concept is useful in training students in this field, as the importance of restoring the function of fingers is equivalent to that of installing a prosthetic hand for the person involved. Experimental results for this device show very good finger performance, with a root mean square error (RMSE) of less than 10−5, showing that it effectively simulates the movement of the finger or parts of the finger, with recognition accuracy of up to 94%.