Simulating pathological gait using Geyer’s neuromuscular model

Details

Supervisors

Ronsse, Renaud

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil biomédical, à finalité spécialisée

Abstract

The ability to walk is essential for both physical and mental health. Disruptive gait disorders can significantly impact a patient’s quality of life. The development of technologies to alleviate these effects is crucial and the field of research is both mature and rapidly evolving. Solutions such as passive and active prostheses or ex- oskeletons can positively enhance the patient’s locomotion, but testing new designs can pose a significant risk if the device-patient interaction produces unexpected behavior leading to falls or injuries. To mitigate this risk, this thesis aims to propose a simulation for pathological gait, creating a safe environment to test and develop new walking assistive aids.This approach is based on Geyer’s neuromusculoskeletal reflex model, which was partially translated from its Simulink implementation to Python by M. Aussems and N. Dineur in their 2023 thesis. This implementation was completed and refined and a reflex adaptation method was developed to tune the biped controller to new musculoskeletal conditions while achieving predetermined metrics. Testing the adaptation on a healthy gait yielded a bipedal model that matched the expected biomechanics of a specific gait, with further analysis on the inner kinetics and kinematics. This method was then applied to model the effects of aging on muscle, testing its ability to predict patient behavior under specific constraints and extract its characteristics to compare to known trends, yielding mixed results. The results of this thesis take steps towards in silico prediction of patient responses to pathology and assistive inputs, thereby reducing the risk of unwanted and dangerous behaviors.