Control of autonomous bicycle system with ZMP tracking

Abstract

This paper proposes zero moment point (ZMP) tracking approach for controlling an autonomous bicycle and improving its stability. In a humanoid robot, ZMP trajectory is widely used to analyze the stability of its locomotion. In the area of stabilizing an autonomous bicycle system, many controllers have been designed, such as PID steering controllers and fuzzy controllers. However, ZMP tracking strategy is first developed in balancing and navigating an autonomous bicycle in considering that ZMP approach is more intuitive and easier to be implemented. The result shows that the performance of the controller with ZMP tracking along straight or circular trajectory is reasonable and conforming to our riding experience.