Computational and Robotic Models of Human Postural Control

Currently, no bipedal robot exhibits fully human-like characteristics in terms of its postural control and movement. Current biped robots move more slowly than humans and are much less stable. Humans utilize a variety of sensory systems to maintain balance, primary among them being the visual, vestibular and proprioceptive systems. A key finding of human postural control experiments has been that the integration of sensory information appears to be dynamically regulated to adapt to changing environmental conditions and the available sensory information, a process referred to as "sensory re-weighting." In contrast, in robotics, the emphasis has been on controlling the location of the center of pressure based on proprioception, with little use of vestibular signals (inertial sensing) and no use of vision. Joint-level PD control with only proprioceptive feedback forms the core of robot standing balance control. More advanced schemes have been proposed but not yet implemented. The multiple sensory sources used by humans to maintain balance allow for more complex sensorimotor strategies not seen in biped robots, and arguably contribute to robust human balance function across a variety of environments and perturbations. Our goal is to replicate this robust human balance behavior in robots.In this work, we review results exploring sensory re-weighting in humans, through a series of experimental protocols, and describe implementations of sensory re-weighting in simulation and on a robot