Modeling And Simulation Of Stable Human Locomotion Using Five Degree Of Freedom Gait Model PDF Download

The human body is a complex system made of more than 600 muscles, which contribute to the actuation of more than 200 Degrees of Freedom (DoFs) [35]. It is therefore a highly redundant system for most kinematic tasks. Many authors have suggested that the central nervous system does not independently control in real-time each muscle and DoFs [54]. Though the high number of muscles and DoFs makes motor control problems difficult, it offers high adaptation capabilities to the body for executing multiple tasks simultaneously when necessary [54]. Among the tasks that require a high level of motor coordination, bipedal gait is a crucial one. The bipedal gait is the natural means of human locomotion. Despite the fact that this movement is quite stereotyped across individuals, it is still unclear how the central nervous system coordinates the complex musculo-skeletal system for gait generation, and how the different sequences of the gait cycle are regulated. In order to address these issues, we proposed to simulate the human-like gait using a simplified model of poly- articulated rigid bodies (3D whole-body skeletal model including 42 degrees of freedom), on which we applied two different motion generation methods. Hence, this thesis is part of the human-like gait generation problem, using motion generation methods from robotics. The first contribution shows that controlling only a small set of adequately selected tasks is sufficient to closely reproduce the human gait kinematics. To this aim, a Hierarchical task controller is applied to the whole-body model with only 3 hierarchical tasks, to generate nine different human-like gaits. The analysis of the simulated gaits shows the emergence of significant human-like properties in walking. In order to validate our results, a comparison between the simulated and human reference joint rotations is conducted. In the end, a discussion is given to illustrate the interest of this approach comparing to related works.The second contribution is based on the well-known hypothesis that human motion is the result of an optimization process. We consider a reduced set of criteria, which seem to be optimized during the human gait, taken from the observation of human walking and the study of the related literature. Direct Optimal Control based on the Differential Dynamic Programming algorithm is applied following these criteria with the whole-body model to generate nine different walking motions. The simulated walking motions are then analyzed and compared to the human reference to show the quality of the gait generation process. The interest of this optimization approach for human-like motion generation is finally discussed. Finally, a comparison between the two methods from robotics is presented and discussed, involving an analysis of the obtained movements' quality.