System identification: theory for the user
System identification: theory for the user
Muscle models: what is gained and what is lost by varying model complexity
Biological Cybernetics
A myosignal-based powered exoskeleton system
IEEE Transactions on Systems, Man, and Cybernetics, Part A: Systems and Humans
| Hi-index | 0.00 |
A crucial issue modelling the relation between the electromyography (EMG) signals and the force produced by a muscle is the general nonlinearity of the plant where standard controller schemes are hardly applicable and the identification of model parameters is complex and tedious. This paper proposes a simple linear process model whose parameters are easy to identify for estimation and quantification of oscillating joint movements like tremor. The model considers the EMG signals of an antagonistic muscle pair acting on the oscillating limb as input and the limb angular acceleration as output. The activation dynamics are modelled through a critically damped 2nd order system, the oscillating movement is considered through a spring-damper system serial to the active part. The spring constant is chosen to be dependent of the limb oscillation frequency. Simulation results with EMG data from four healthy subjects performing tremor-like wrist oscillations are given and the performance of the proposed model structure is compared to a linear model not considering the specific movement and an underdamped model with fixed parameters.