Fig. 3
From: Using Bayesian inference to estimate plausible muscle forces in musculoskeletal models
Generating muscle excitations from compact radial basis functions (CRBFs): The excitation signal for each muscle was determined by the amplitudes of ten CRBFs, each with a pre-defined center and width. A One CRBF shown individually. C All ten CRBFs with preset centers and widths, with varied amplitudes. E The sum across all ten CRBFs for a given muscle. G The summed output is then converted via an inverse-logit transform (Eq. 2) so that the excitation is constrained to be between 0 and 1. These excitations are then sent to the OpenSim model to perform the forward integration. The right column B, D, F and H shows the effect of changing the amplitude of a single CRBF (black dashed line). Changing the amplitude of the CRBF affects the muscle excitation signal within the width of the CRBF, but the area outside of the CRBF remains constant