Jogo sério baseado em realidade virtual associado ao processamento de sinal EMG e feedback sensorial para treinamento de prótese de membro superior

Abstract

Amputation is a lasting life experience. Furthermore, prosthesis fitting and training are usually long and sometimes frustrating experiences. Few approaches explore the problem of controlling virtual myoelectric prostheses, using Virtual Reality techniques, as it is not a trivial issue. Besides, the lack of proper sensory feedback, when the system warns the amputee that there has been contact with objects during training, generally leads to a very “artificial” experience for most users. In this work, we propose a system that uses Immersive Virtual Reality (IVR) and EMG (signals that measure muscle activity) signal processing to provide a training environment for amputees who are supposed to use myoelectric prostheses. Here, we also investigate the efficiency of learning how to control a virtual prosthesis with and without sensory feedback. The system runs on a Vive Pro equipment using a Vive Tracker to track the user's arm position and orientation, and mirror the movements into a virtual prosthesis. EMG signals control the opening and closing of the virtual prosthesis, vibration motors placed on the user's forearm provide sensory feedback. The results show that virtual training can be significantly improved when proper tactile feedback is provided, especially for the myoelectric control system. The use of elements from serious Games provided an interactive and specific environment for training volunteers. Both EMG control and tactile feedback have provided an effective control loop allowing for a better sense of realism for the user’s training protocol.