Can a Passive Dynamic Walking Robot Exhibit a Deterministic Nonlinear Gait?

There is a growing body of evidence that nonlinear step-to-step variations are related to how the nervous system controls stable locomotion. However, we still have a limited understanding of how the neuromechanical variables of the locomotive system influence the observed nonlinear structure. Reliable models that closely resemble the nonlinear gait dynamics are required to advance our knowledge in this scientific field. Recently our lab and others have demonstrated that a passive dynamic walking model can have a deterministic nonlinear gait pattern (Garcia et al., 1998; Kurz & Stergiou, 2006). Results from the simulations of these models suggest that it may be possible to build a physical walking robot that has a deterministic nonlinear gait pattern. Here we explored if the results from our simulations can be extended to a physical passive dynamic walking robot, which can provide insight into the nonlinear gait patterns observed in human walking.