A mathematical analogue of electrically stimulated muscle using local model networks

The authors are searching for mathematical models, which can represent the complex nonlinear behaviour of muscles, so that the contractions can be better controlled. The authors have developed a number of dynamic muscle models based on force measurements taken from a muscle stimulated by an irregular pulse train. The results of experimental investigations have shown that a simple linear transfer function model of muscle response can be rather inaccurate over the full operational range. The muscle response is significantly nonlinear with respect to the value of isometric tension (the output response to pulse stimulation) and it is also known that muscle fatigue over time results in significant dynamic variations. Nonlinear models based on local model networks (LMN) are able to capture the nonlinear effects and provide accuracy over a wide operational range. The authors describe the type of network model used, the training algorithm, and show some typical experimental results based on real measured data