Anmerkungen:
Diese Datenquelle enthält auch Bestandsnachweise, die nicht zu einem Volltext führen.
Beschreibung:
The myoelectric control of hand prosthesis commercially available is simple and limits the user to very basic operations. Although in the academic research for prosthesis control a large variety of advanced control methods has been developed, none of them has replaced the current industrial state of the art, yet. In this PhD project I have investigated and developed an approach towards intuitive prostheses control, based on new signal-processing and regression algorithms. By introducing a novel adaptive pre-processing algorithm (ACAR) for the surface EMG signals and designing a regression system based on a non-negative matrix factorization, I have developed a myocontrol system capable of online control of upper limb prosthesis for two degrees of freedom, simultaneously and proportionally. Additionally, I have developed a virtual evaluation paradigm, which can assess the control performance of important hand movements necessary for daily life activities. This online assessment goes beyond the state of the art of myoelectric control research, which is done offline. That is without the interaction with the subject. The resulting myocontrol system and virtual evaluation paradigm have been tested in both intact-limb subjects and subjects with limb deficiencies. In these studies, the benefits of the developed algorithms have been confirmed. The scientific results and developments of this project have been the basis for additional publications and scientific achievements by the Department of Neurorehabilitation Engineering and its scientific partners. This underlines the impact of this work in the field of myoelectric control for upper limb prostheses.