direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Logo der TU Berlin

Inhalt des Dokuments

Nonlinear Modelling, Identification and Robust Control of Electrically Stimulated Muscles

From Fachgebiet Regelungssysteme TU Berlin

Jump to: navigation, search

Contents

Abstract

This research addresses topics related to reduced modelling, parameter identification and robust control of electrically stimulated muscles.

Cooperation

People involved

Description

Neuro-musculoskeletal systems are complex, nonlinear and, due to muscle fatigue, rapidly time-varying. An adaptive scheme based on on-line identification of simplified model structures is therefore an attractive possibility for controller design. In this context, restrictions imposed by the intended use in a clinical environment are of prime importance. For example, goniometer measurements of joint angles are to be preferred to powerful optical analysis systems, simply because the latter may be affected by physiotherapists working with the patient. The identification scheme proposed in [1] takes such restrictions into account. An implicit self-tuning scheme, which does not need explicit identification of a plant model, has also been investigated and is described in [2]. Another approach is to work with less detailed model information, to emphasise robustness during controller design and to accept a certain degree of performance degradation. We have applied backstepping control synthesis techniques to a lowerlimb model reflecting the well-known equations of motion and passive joint properties, but containing only an extremely rudimentary description of muscle dynamics plus the corresponding uncertainty bounds [1].

Publications

  1. 1.0 1.1
    Thomas Schauer, Nils-Otto Neg\aard, Fabio Previdi, Ken J. Hunt, Mathew H. Fraser, Elke Ferchland, Jörg Raisch. Online Identification and Nonlinear Control of the Electrically Stimulated Quadriceps Muscle. Control Engineering Practice, 13 (9):1207–1219, 2005.
  2. Fabio Previdi, Thomas Schauer, Sergio M. Savaresi, Ken J. Hunt. Data-Driven Control Design for Neuroprotheses: A Virtual Reference Feedback Tuning (VRFT) Approach. IEEE Transactions on Control System Technology, 12 (1):176–182, 2004.

Recommend this page