Analyses of muscle force predictions based on optimization

Static-optimization is one method of estimating individual muscle forces produced during a given movement. In the past, the resulting optimization problem as been solved analytically to study specific properties of individual muscle force predictions when using a non-linear optimization algorithm. However, no framework for analyzing the general properties of muscle force predictions is available. The purpose of this thesis was to develop and apply a general framework for analyzing individual muscle force predictions using non-linear optimization. We show here that non-linear optimization can predict experimentally observed force-sharing loops and that there is a “hyper-plane ” that separates the moment-arm vectors of active and passive muscles. It is also shown that this approach is valid for both planar and three-dimensional systems. This thesis comprises a purely theoretical analysis of the force-sharing problem in biomechanics. Future work should include developing specific experiments to test the general force-sharing predictions found here, theoretically.