Polytopic uncertainties identification for electrically stimulated lower limbs

Abstract

Functional electrical stimulation (FES) has been used for a wide range of applications in individuals with spinal cord injury. Spinal cord injury rehabilitation modalities based on FES include gait training programs, restoration human-body vital function, and others. The various control methodologies require a system mathematical model, therefore, it is necessary to identify the parameters of this system. It is an extremely complex challenge to control the knee angle or the torque shank-foot complex through the functional electrical stimulation applied to the muscle. The response of the stimulated muscle is nonlinear and time-varying, allowing the fatigue occurrence due to muscle fibers non-physiological recruitment and altered muscle composition caused by FES. Relevant studies show that system developed in closed-loop satisfactorily deals with disturbances and uncertainties of the plant to be controlled. The contribution of this paper is to propose an analysis about polytopic uncertainties identification of the electrically stimulated lower limbs model. Knowing a priori the uncertain parameters of the linearized plant, the performance of a possible controller could be more efficient. In future works, it will be possible to design robust control techniques via linear matrix inequalities. Through this study, it is possible to work a large sample of volunteers with the objective of measuring muscular behavior when using the FES application.