Generation of Large Gymnastic Motions for a Family of Under-Actuated Robots by Zero-Moment-Manifold Control
Hiroe Hashiguchi, Suguru Arimoto, Toyohide Sakamoto, Shoji Namikawa
- Year
- 2006
- Citations
- 2
Abstract
This paper is concerned with a novel control method to generate any desired motion for a family of under-actuated robots such as gymnastics robots and acrobots having a single lower leg pivoted at the floor. This pivotal ankle joint can not be actuator-driven but may have only a passive viscous device such as a rotary damper. Therefore these robots are intrinsically unstable under the effect of gravity. In order to maneuver such an under-actuated robot with n joints under non-holonomic constraint without loosing stability of motion, we have proposed a zero moment manifold (ZMM) control method by introducing an (n - 1)-dimensional configuration manifold, on which a set of joint angle vectors satisfies zero of the rotational moment around the ankle joint. It could be shown theoretically that any motion starting from any initial posture on a subset of the ZMM and moving to a desired posture can be stabilized by using a coordinated control composed of gravity compensation and PD feedback. Though the theoretical proof can guarantee only stability of movements of joint angles in a narrow range, it has been observed that the robot can move in a wider range through preliminary numerical simulation results. Notwithstanding such a conservative result of theory, this paper shows that the ZMM control with constant PD gains can generate global gymnastics movements for a class of under-actuated robots by verifying through a variety of numerical simulations that most of given attractors (target postures) have corresponding large attraction regions. This is also verified experimentally by using a real gymnastics robot with four joints and a series of large movements can be produced consecutively as programmed by selecting several target postures enough separated each other and combining them
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991