Abstract:
For decades, biologists believed that animals walk because the brain calculates motion trajectories for their limbs. However, an alternative hypothesis on animal locomotion suggests the opposite. The animal framework is built to walk “naturally” and the dynamics work without relying on controls. Instead, the walking gait is simply generated by the interaction of gravity and inertia, establishing a stable, naturally emerging limit-cycle known as passive dynamic walking. The feasibility of passive dynamic walking has been demonstrated for a biped system consisting of only a pair of legs. This thesis examines full-body passive dynamic walking models with simple and animal-like mechanical linkages that can generate walking gaits using only gravity that are able to recover from small perturbations without the need for controller input. The contribution of a torso to the stability and efficiency of passive biped walking is also addressed. When an upper-body is added, passive dynamic walking takes place on level ground but for energetic reasons, it is unstable. The second part of the thesis looks at how to stabilize the passive walking trajectory on level ground in a physically feasible and biologically relevant way. Findings suggest that the role of locomotion control is to provide stability, rather than drive the limb onto a pre-calculated trajectory.
