Estimating human joint torques
and muscle activations:
A new approach to inverse dynamics
without ground reaction data

Abstract

The primary aim of this thesis is to develop a novel inverse dynamics method to evaluate joint torques and muscle forces for analysing human gait. Classical inverse dynamics methods rely on the direct measurement of ground reaction forces (GRFs), which still faces major challenges in various environments, particularly outside controlled laboratory settings. While advancements in kinematic measurements have facilitated the acquisition of motion data in diverse contexts, the reliance on expensive force platforms restricts the scope of biomechanical studies.

The objective of this thesis is to overcome these limitations and eliminate the need for force platform data in the context of inverse dynamics analysis. The research led to the creation of a three-dimensional contact model based on fundamental assumptions about foot-ground interaction, which was later used in a complete biomechanical analysis workflow in order to estimate joints torques and muscle activation.

Jury members  

• O. BRULS, Université de Liège, Belgium (Advisor)
• L. GERIS, Université de Liège, Belgium (President)
• V. DENOËL, Université de Liège
• C. SCHWARTZ, Université de Liège
• J-L. CROISIER 
• P. FISETTE, Université catholique de Louvain
• A. KECSKEMETHY, Universität Duisburg-Essen (Allemagne)

Videoconference available on Teams