Model-based bounding on a quadruped robot
Chun-Kai Huang, Chung-Li Chen, Chia-Jui Hu, Pei‐Chun Lin
- 发表年份
- 2016
- 引用次数
- 6
摘要
We report on the development of a robot's dynamic bounding locomotion based on a sagittal plane model that fits the robot's natural dynamics. The proposed model, referred to as the Two-rolling-leg (TRL) model, is a low degree-of-freedom planar rigid-body model with two rolling and compliant legs. The numerical fixed point analyses of the model suggest that with the adequate touchdown conditions of the model, passive dynamic and periodic behaviors of bounding is achievable. The passive dynamic behaviors of the model then serve as a guidance to initiate bounding on a quadruped. The experimental results show that though several un-modeled factors cause the behavioral discrepancy between the TRL model and the robot, the robot can still initiate its dynamic bounding behaviors by mapping the corresponding leg trajectories of the model on the empirical robot using a simple pre-set position control strategy without any state feedback. This work confirms that the complex gait behavior of the robot can be initiated by adopting the simple reduced-order model as the control target for behavioral guidance.
关键词
相关论文
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991
A new optimizer using particle swarm theory
R.C. Eberhart, James Kennedy
2002