Be Careful What You Wish for: Cost Function Sensitivity in Predictive Simulations for Assistive Device Design

Software packages that use optimization to predict the motion of dynamic systems are powerful tools for studying human movement. These “predictive simulations” gaining popularity in parameter studies designing assistive devices such as exoskeletons. The cost function is a critical component problem and can dramatically affect solution. Many functions have been proposed biologically inspired produce reasonable solutions, but which may lead different conclusions some contexts. We used OpenSim Moco generate predictive simulations walking using several functions, each produced trajectory model. then augmented model with motors generated hip flexion, knee or ankle plantarflexion torques, repeated determine optimal motor torques. was assumed be planar bilaterally symmetric reduce computation time. Peak torques varied from 41.3 79.0 N·m flexion motors, 48.0 94.2 42.6 79.8 could important design consequences. This study highlights importance evaluating robustness results simulations.