A nonlinear parametric identification method for biceps muscle model by using iterative learning approach

This paper focuses on the modeling of the human bicep brachii muscle and introduces an iterative identification method for nonlinear parameters in a virtual muscle model. This model displays characteristics that are highly nonlinear and dynamical in nature. However, the precision of the virtual muscle model depends on a set of model parameters which cannot be acquired easily using non-invasive measurement technology. Hence, experiments were conducted to derive relationships between joint angles, force, and EMG signals. In the experiment, the calculations from an anatomical mechanical model were used to relate isometric force to EMG levels at 5 different elbow angles for 3 subjects. An iterative identification method was then used to determine optimum muscle length and muscle mass of the biceps muscle based on the model and muscle data. Extensive studies have shown that the iterative identification method can achieve satisfactory results.