Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Solving optimal control problems applied to neuromusculoskeletal models: a comparison between OpenSim Moco and Fmincon Solvers
Submission Author:
Denis César Mosconi Pereira , SP
Co-Authors:
Denis César Mosconi Pereira, Adriano Siqueira
Presenter: Adriano Siqueira
doi://10.26678/ABCM.COBEM2023.COB2023-0120
Abstract
Optimal control problems applied to neuromusculoskeletal models are useful to allow predictive simulations in the field of biomechanics. In this context, a challenge that biomechanical engineers face is the formulation and solution of such problems, as even with several solvers available, writing algorithms is not always simple or the results are not always satisfactory, given the limitations of the solver. The objective of this work was to compare the use of OpenSim Moco and Fmincon to solve an optimal control problem applied to a neuromusculoskeletal model of a human leg performing an open kinematic chain movement. The results obtained proved that both solvers are able to lead well with human neuromusculoskeletal models, providing physically possible solutions on human torque, muscle activations and movement, being useful for the understanding of human movement as well as the development of devices and protocols related to the biomechanical field. When comparing the usability of the two solvers, the OpenSim Moco proved to be easier to use in addition to being fast and effective in solving the problem, while Fmincon, despite being reasonably more complex and slower, proved to be more flexible.
Keywords
OpenSim Moco, Biomechanical simulation, direct collocation
