Adjusting Legs for Stable Running in Three Dimensions

Abstract

Spring-like leg behavior can be found in human and animal locomotion both in sagittal and horizontal plane. Based on this observation, two-dimensional models with spring-like leg function were developed which predict self-stability for appropriate leg parameter adjustments. Hence, system perturbations can be compensated already on the mechanical level reducing the need for feedback control. So far, the extension of these models to three dimensions could not confirm the self-stabilizing effect of running with spring-like legs. Therefore, we investigate how self-stability could be achieved in 3D running. For this, we analyze the effects of different leg placement protocols on the overall running behavior. We adjust the leg laterally with respect to the velocity vector and investigate two different leg placement protocols for vertical leg alignment taking into account velocity vector and vertical axis. Both investigated leg placement protocols show self-stable running patterns with no minimal limit speed which is in contrast to two-dimensional spring-mass models. Furthermore, self-stable transitions from hopping to running and vice versa were found, induced by adjusted leg placement only.