Application of multiple model adaptive control to upper limb stroke rehabilitation

Abstract

Impaired arm function has a significant impact on the quality of life of stroke sufferers. Research has shown that the application of functional electrical stimulation (FES) to assist their movement over repeated attempts at a task is effective in restoring function. However, current FES control systems lack robustness to changes in plant dynamics caused by fatigue and spasticity. This paper details the application of a multiple model adaptive control algorithm that has the potential to overcome this problem. It is shown in an experimental setting that an adaptive estimation mechanism is able to detect changes in the true plant through optimal disturbance estimation. Finally, the performance of the algorithm is compared with that of fixed optimal controllers. These initial results suggest that the application of estimation-based multiple model switched adaptive control (EMMSAC) can increase the potential of FES-based rehabilitation through improved controller accuracy.