Subproject C01
Subproject C01
Machine Learning for Personalization of Musculoskeletal Models, Movement Analysis, and Movement Predictions
The extent to which a neural network can be used to effectively personalize gait simulations using motion data is explored. In this subproject, we first investigate the influence of body parameters on gait simulation. An initial version of the personalization is trained with simulated motion data, since ground truth data is known for this purpose. We then explore gradient-free methods to fit the network for experimental motion data. The resulting network is validated with magnetic resonance imaging, electromyography, and intra-body variables.
The optimal movement of a person is measured through different criteria like cost of transport, muscular effort, impact, head stabilization, and others. Inverse Optimal Control (IOC) is used to best recreate specific motions by combining weighted criteria. To explore human walking criteria like the influence of environmental and personal conditions we are developing a fast IOC pipeline for real-world data and predicting new gaits from weighted criteria
Contacts