Human Locomotion: How Humans Move Efficiently and Stably

Healthy humans walk and run with lower energy consumption than most bipedal robots. Humans also move with greater stability, robustness, and versatility. In this two-part talk, I will first describe our experiments and optimization-based predictions, demonstrating that energy optimality can predict many aspects of human locomotion behavior: steady locomotion in a straight line, unsteady locomotion with changing speeds, and non-straight-line locomotion. Building on this evidence, I will describe a computational framework for designing robotic prosthesis and exoskeletons that minimize human walking effort, discussing two ongoing studies. In the second part of the talk, I will describe our attempts to characterize the controller humans use to walk and run stably. We performed human experiments to record responses to perturbations and deviations from normal locomotion, and derived simple controllers that fit these human control responses. We show that some aspects of these controllers can also be explained, at least qualitatively, by energy-optimal perturbation recovery.