Predictive Simulation of Sit-to-Stand Movements

The development of technology that can assist people during movements such as sit to stand can benefit from simulations that can estimate the effect of different assistance patterns. These simulations need to be based on frameworks that can replicate the neuromechanical features of non-pathological sit-to-stand movements. In this study, we present a framework for predictive simulations of sit to stand that can be used to synthesize physiological data in the absence of experimental ones. Here we reproduce sit-to-stand movements on a model with 9 degrees of freedom and 52 actuators by optimizing the parameters of a feedback controller that accounts for the position, velocity and acceleration of the pelvis and torso, under the constraints dictated by a specifically designed cost function. We obtained a simulation that could replicate the kinematics and muscular activation observed in previous studies.