VBLA, a swing leg control approach for humans and robots

Abstract

Experiments on human subjects, data analyses and modeling can help the engineers design and develop high performance humanoid robots and assistive devices. In an abstract level bipedal locomotion can be considered as a combination of three sub-functions: stance, swing and posture control. In this paper, we focus on swing leg adjustment searching for a bio-inspired method working well on simulation models and robots. In that respect, velocity based leg adjustment method (VBLA) is presented to find the desired leg angle in different gaits. Our investigations are based on analyses of human walking, perturbed hopping experiments, beside simulation studies on bipedal running and walking. Compared to some other approaches the VBLA can better explain human leg adjustment in different gaits, gives higher robustness against parameter variations and is practical and easy to implement on robots.