Humans use visual, vestibular (gravito-inertial), and proprioceptive (sensation of joint movements) sensory information to maintain an upright position. In the last 15 years systems identification techniques have been increasingly applied to understand how humans integrate these sensory cues to generate context dependent torque in face of the multitude of changing environmental conditions.
Based on such techniques and extensive research on human perception, our group at the Neurology department has developed a control model (called Disturbance Estimation and Compensation model) that is able to quantitatively reproduce a wide range of human sway characteristics in standing balance. The talk will focus on the sway responses to pseudo-random support surface tilt motions in humans, model simulations and robot experiments. It will highlight how the human control mechanism deals with the loss of sensory information, properties of the nervous system such as long time delays (>100 ms), sensory noise, etc., and will point out some difficulties for systems identification resulting from the non-linear nature of the mechanism.
Presenter: Lorenz Assländer has studied Physics and Sport Science and did his PhD on the integration of visual cues in human standing balance at the Neurocenter Freiburg. The main focus of his research is human motor control, and specifically the model based understanding of sensory integration in standing balance.