Computational models of upper-limb motion during functional reaching tasks for application in FES-based stroke rehabilitation

摘要: Functional electrical stimulation (FES) has been shown to be an effective approach upper-limb stroke rehabilitation, where it is used assist arm and shoulder motion. Model-based FES controllers have recently confirmed significant potential improve accuracy of functional reaching tasks, but they typically require a reference trajectory track. Few control schemes embed computational model the task; however, this critical ensure controller reinforces intended movement with high accuracy. This paper derives motor models tasks that can directly embedded in real-time schemes, removing need for predefined trajectory. Dynamic electrically stimulated are first derived, constrained optimisation problems formulated encapsulate common activities daily living. These solved using iterative algorithms, results compared kinematic data from 12 subjects found fit closely (mean fitting between 63.2% 84.0%). The performed iteratively variables hence transformed into learning algorithm by replacing simulation signals experimental data. therefore capable controlling real time manner corresponding unimpaired natural movement. By ensuring assistance aligned voluntary intention, maximises effectiveness future rehabilitation trials.