A lower limb EMG-driven biomechanical model for applications in rehabilitation robotics

Current changes in aging demographics poses new challenges: people require to keep their quality of life even after circumstances that threatened their movement and function. This increases the demand for new physical rehabilitation facilities that go beyond the traditional patient-therapist, one-to-one rehabilitation sessions. Two promising solutions rely on virtual reality and on the development of autonomous active orthoses, or exoskeletons. Whatever is the chosen approach, there is a requirement for a robust human-machine interface for the control, able to understand patient's intention and to produce an immediate activation of the device. This paper presents a biomechanical model, a possible solution able to predict joint torque from the surface electromyography signals emitted by muscles during their activation. The main objective of the research is to investigate the benefits and efficacy of this model and to lay down the basis of our current research, whose main goal is to make possible a rehabilitation process either with active orthoses or virtual reality. Experiments involving all the steps of our model demonstrate the viability and effectiveness of our approach.