Dynamic modeling and balance control of human/bicycle systems

Abstract

Understanding of human/bicycle interaction is crucial not only for enhancing human/bicycle safety but also for designing new bicycle-based robotic device. We present a dynamic modeling and balance control of human/bicycle systems. The dynamic models of human/bicycle systems are based on previously developed modeling of riderless autonomous motorcycles. We consider the human body as a point-mass connected to the bicycle platform. For balancing control of human/bicycle systems, we propose a hierarchical sliding-mode control approach in which multiple coupled sliding surfaces are designed. The human body action is considered as an unknown disturbance and a nonlinear disturbance observer (NDOB) is used to estimate its value. The stability of the combined control and NDOB systems is guaranteed. The performance of the control systems is demonstrated through simulation results.