Stochastic optimal control simulations of human walking for computer-aided exoskeleton design

Predictive simulations allow for the generation of novel movements based on neuro-musculoskeletal models. Hence, predictive simulations can be used to study the effect of changes in the musculoskeletal system or motor control strategy (e.g. due to aging, pathology or treatment) on movement patterns. Human movement is robust against internal (sensorimotor noise) and external perturbations. Yet, simulation frameworks that describe this robustness are computationally expensive and have therefore only been applied based on simple musculoskeletal models. The aim of this PhD project is to develop computationally efficient methods to simulate movement in the presence of noise and to apply these methods to simulate human locomotion based on complex musculoskeletal models. Simulations will be combined with experimental data to advance our insight in how age-related and pathological (e.g. cerebral palsy) decline in sensorimotor acuity influences walking balance. Next, we will provide proof of concept that the simulation framework can improve the design of interventions by using it to design orthoses that improve walking performance in children with cerebral palsy.