Balance equilibrium manifold and control of rider-bikebot systems

Postural balance control of machines is one of the most important human motor skills. We present a balance equilibrium manifold (BEM) concept to study how a human rider balances a bicycle-like robot (i.e., bikebot) while maintaining tracking a desired trajectory. The BEM is built on the external-internal convertible (EIC) structure of the rider-bikebot dynamics to capture the balance properties of the rider-bikebot systems. We analyze the contribution of the upper-body movements and the steering inputs to the task of balance control of the systems. The results reveal that steering actuation is more effective than upper-body movement for balance task. A performance metric is also introduced to quantify the balance motor skills using the BEM. Extensive experiments are conducted to validate and demonstrate the analyses and the proposed balance skill metrics.