Grasp and rotate fragile objects using fuzzy control and tactile sensor

Last years a lot of research work has been devoted in the area of prosthetic hands towards achieving behaviour similar to that human hand. These prosthetic hand allow complicated operations such as grasping and manipulating of fragile objects to be realised. For the implementation of such prostheses, the patient receives signals coming from the environment, in order to allow an approximation of the sense of touch. In addition, sophisticated control systems, electronics and actuation have to be proposed, in order to obtain a better performance of the prosthetic hand. In this work, a new controller for artificial hands with two fingers used to grasp and rotate fragile and delicate objects is proposed. In this controller tactile feedback is used and the closing velocity of the fingers is adjusted, in order to grasp the object by applying the minimal required force and without the measurement of this force. The operation of the proposed controller is based on fuzzy logic for the following two reasons. The first is the ambition to achieve prosthetic hand behaviour similar to that of a human hand when it holds an object. The second reason is that using fuzzy logic the indispensable mathematics can be simplified in order to design a system operating in real time. Finally, the results from the comparison of the proposed controller efficiency with PID's are presented.