This project contains code and data to perform 3D muscle-driven predictive simulations of gait.
Predictive simulations of human movement have great potential but are often limited by large computational costs. In this project, we developed an OpenSim-based framework to perform computationally efficient predictive simulations of movement.
The framework relies on numerical tools including direct collocation, implicit differential equations, and algorithmic differentiation, and generates predictive simulations of gait in about 35 minutes (single core of a standard laptop computer) with muscle-driven 3D models (29 degrees of freedom and 92 muscles).
The code contains a series of example predictive simulations in which we varied objective function, musculoskeletal properties, and gait speed. Details of the results can be found in the associated publication: Falisse et al, Rapid predictive simulations with complex musculoskeletal models suggest that diverse healthy and pathological human gaits can emerge from similar control strategies, Journal of the Royal Society Interface (2019), in press.
The released folder contains readme files with details of the folder structure. Please contact me for any questions (firstname.lastname@example.org) or post an issue on GitHub https://github.com/antoinefalisse/3dpredictsim.