Stable control of the bicycle robot on a curved path by using a reaction wheel

• Published in: Journal of mechanical science and technology Vol. 29; no. 5; pp. 2219 - 2226
• Main Authors: Kim, Yunki, Kim, Hyunwoo, Lee, Jangmyung
• Format: Journal Article
• Language: English
• Published: Seoul Korean Society of Mechanical Engineers 01.05.2015; Springer Nature B.V; 대한기계학회
• Subjects: Acceleration; Algorithms; Bicycles; Centrifugal force; Control; Control algorithms; Control stability; Control systems design; Control theory; Controllers; Dynamical Systems; Engineering; Floors; Friction; Industrial and Production Engineering; Mechanical Engineering; Proportional integral derivative; Reaction wheels; Robot control; Robots; Roll; Rolls; Steering; Vibration; 기계공학; Process of controlling the balance; Centrifugal force; Friction cone; Reaction wheel; S-curved motion; Bicycle robot
• ISSN: 1738-494X; 1976-3824
• DOI: 10.1007/s12206-015-0442-1

A new velocity and balance control algorithm for the RWBR (Reaction wheel bicycle robot) has been proposed in this paper. A reaction wheel is adopted to maintain a balance while the RWBR is driving, which allows the process of controlling the speed to be achieved concurrently with the one of controlling the balance. To control the speed of the bicycle robot, a PID control algorithm with the adjustment of variable gains is developed in this study, where the gains are heuristically adjusted during each experiment. For the control of the balance and stability, a roll controller is designed by using the model-based algorithm to provide the shortest possible cycle for the bicycle controller. The roll angle is measured to maintain the desired acceleration which generates the reaction force to keep the total force acting on the bicycle robot inside the friction cone. The desired roll acceleration is decided for the bicycle robot in order to prevent it from falling down to the floor with the minimum usage of the reaction wheel rotation. For the general driving of the bicycle robot such as the S-curved driving, the dynamic forces of centrifugal, gravity and steering change are modeled and the friction on the floor has been estimated. The performance of the designed control system is verified through the real experiments with the developed RWBR.