Use this URL to cite or link to this record in EThOS:
Title: Implementation and experimental evaluation of multiple model switched adaptive control for FES-based rehabilitation
Author: Brend, O.
ISNI:       0000 0004 5359 2134
Awarding Body: University of Southampton
Current Institution: University of Southampton
Date of Award: 2014
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Functional electrical stimulation (FES) is a well-established approach that is employed as a therapeutic tool for the restoration of motor control in individuals experiencing muscle impairment. Although its use as a rehabilitation tool is validated by clinical results, current control approaches limit the full exploitation of its potential due to the lack of accuracy with which the FES is applied. Research has thus focused on the use of advanced, closed-loop control algorithms to provide more accurate FES that is both task-oriented, and matches the rehabilitation needs of the patient. Experimental results have been reported for a variety of control schemes. However, the majority of approaches have failed to transfer to clinical practice due to the difficulties associated with identifying a model of electrically stimulated muscle that adapts as the true plant varies with time. Estimation-based multiple model switched adaptive control (EMMSAC) is a robust control approach that has the potential to overcome the problems associated with the uncertain, time-varying properties of electrically stimulated muscle. EMMSAC utilises optimal disturbance estimation to assess the respective performances of a set of candidate plant models. Then the controller associated with the model that has best performance is switched into closed-loop operation. This thesis details the algorithmic modifications that allow disturbance estimation to be performed in the time-varying setting for nonlinear Hammerstein structures. Then it is shown experimentally that a general plant model set can be identified that represents the time-varying, FES-induced muscle activation dynamics for the population of younger healthy adults. This finding is exploited to design an EMMSAC controller that achieves accurate trajectory tracking for multiple participants with minimal prior model identification. Results indicate that the use of EMMSAC reduces RMS tracking error when compared with a fixed controller; similar results are also reported for older healthy participants. Furthermore, initial results for a small sample of stroke-participants are shown, which confirms the potential for the proposed control approach to be applied in a clinical setting for FES-based rehabilitation.
Supervisor: Freeman, Christopher Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: R Medicine (General) ; TK Electrical engineering. Electronics Nuclear engineering