Automated Optimal Model Calibration for the OpenSim Biomechanics Simulator
Thomas Uchida (University of Ottawa)
To develop an automated tool that uses optimization to calibrate models of the musculoskeletal system and improve simulation results, and disseminate the tool to the OpenSim community with documentation, examples, case studies, and outreach events.
OpenSim is an open source software platform for modeling musculoskeletal systems and generating simulations of human movement. Simulations that reproduce recorded movements enable estimation of variables that cannot be measured directly, such as joint loads and tendon strains, and have enabled discovery of new ways to prevent injury, treat disease, and enhance performance. Most OpenSim users are researchers who spend months learning the software. To make OpenSim’s capabilities available to a broader community of engineers and clinicians, the modeling and simulation process must be simpler. This project will automate model calibration, or the procedure by which a generic model is adjusted to match the dimensions of a human subject, to improve the process for new users.