Sensory Augmentation in a Complex Balance Task:

Balance rehabilitation is of considerable interest. Vibrotactile feedback offers an intuitive pathway for postural and mobility decisions. However, measuring the performance improvements of any feedback in conventional balance suffers from a general problem of decoupled system and measurement – thus ceiling effects are common and it is difficult to accurately measure and predict performance.

Human balance is complex and multi-faceted relying on different sensory systems and control strategies depending on the task and environment. The assessment of human balance is also complex, relying on inference from sensory and functional tasks. Posturography is commonly used as a quantifiable measurement of stance (or gait). However, the healthy human balance system is relatively stable during typical stance and most ambulation tasks. Human balance is therefore typically a multiply redundant, decoupled feedback system that exhibits a high margin of stability. It is therefore very difficult to measure the stability margins and feedback characteristics of the human balance system.

Unicycles present an unusual control system problem; much like the inverted pendulum control experiment they have a singularity at the zero point. Thus the control system must be continuously performing (over a tight range) for the system to remain upright and stable. Further, control of the unicycle is non-linear requiring the combination of several actions to affect motion. Mechatronics has recognized these factors and used the unicycle from which to develop new robust nonlinear control strategies such as disturbance observer techniques etc.

Our research question is whether the human control strategies used in riding a unicycle can be analogous to the more general problem of balance control. Specifically, can vibrotactile feedback to unicycle riders improve performance?