Intelligent Bio Robotic Arm For Mapping Signal Control And Synchronisation For Motor Control System Using Human Computer Interface.
Main Article Content
Abstract
Humans rely so heavily on their hands to do everyday activities, communicate, and show emotion that losing a hand might make you feel as though you have lost your entire sense of independence. There are several excellent prosthetic hands available today that offer extraordinary movement flexibility. Because they are difficult to operate, they aren't used to their full potential. They can almost match the whole range of motion of the natural hand. In the proposed model, we are aiming to achieve natural intuitive control of prosthetic hands by using electrical activity in the surviving muscles to send control signals to the motors of the prosthetic hands. However, the number of signals you can record is constrained, limiting the movements you can make. In addition, the lack of feeling or feedback provided to the user results in unnatural control, as users lack the sense of touch and force on the hands. This suggests that controlling the gadgets requires a great deal of concentration and is not especially natural. Therefore, to enhance prosthetic hand control, we use machine learning modelling with smart cameras. From the recorded muscle signals, artificial intelligence may infer how the missing hand would naturally move. These movement commands can then be processed using real-time object detection and sent to the prosthetic hand. This method has the benefit that by considering information about the biomechanics of the hand, we can predict what users are trying to accomplish more accurately. Additionally, the model created is straightforward, so there is no need for special mental training to carry out simple activities. The accuracy of object grasping varies from 65% (Minimum) to 85 % (maximum). The cost of this proposed model is less as compared to the models available in the market or the researchers proposed.