Iterative learning control for robotic-assisted upper limb stroke rehabilitation in the presence of muscle fatigue

The use of iterative learning control to regulate assistive functional electrical stimulation applied to the muscles of patients undergoing robotic-assisted upper limb stroke rehabilitation has been followed through small scale clinical trials. These trials confirmed that an increase in patient ability to complete the specified task also led to a reduction in the level of electrical stimulation required. This previous work assumed that the effects of muscle fatigue could be neglected but if a patient suffers fatigue during a rehabilitation session then their the session goals are not achieved or, more likely, the session must be abandoned due to the time limits imposed by the ethical approval required to conduct such sessions. In this paper the results of the first investigation into enhancing the control scheme to remove or lessen the effects of fatigue and hence make better use of the time available for a session are given. The scheme considered adds a feedback loop around the muscle model used, where the performance results given are based on a model for the dynamics constructed using patient data collected in previous clinical trials.