Knowledge of the movement and internal forces of the human body is essential in many areas of human medicine and in particular in orthopedic and traumatological surgery. Modern numerical methods are used to estimate forces within body structures from recorded motion and external loads. However, their general utilities are limited by their inability to predict movement and internal charges in response to changes caused by a pathology or as a result of the application of a treatment to treat a pathology.

The aim of this thesis project is to study and design numerical models of the human body making possible the representation of musculoskeletal disorders in humans. It is also a question of proposing and developing algorithmic methods allowing to predict the walking movement. This work will result in the development of a new generation of predictive simulator, based on the calculation of optimal movements, with the objective of being able to predict and thus improve the therapeutic strategies specific to a patient. The simulator will propose numerical means to guide and predict therapies, and may complement current empirical approaches that rely on the evolution of therapies based on measured data. The work will be inspired by direct simulation methods that have so far been developed primarily for the representation of human movement in the entertainment and gaming industry. An innovative predictive musculoskeletal simulator will be developed and implemented for this purpose.

The objectives of the thesis are therefore (i) to develop a musculoskeletal motion controller for human locomotion, (ii) to model the targeted disorders and (iii) to test and validate the simulator in a clinical setting.