PR-MPC for NAO H21 Humanoid Robot: Capture Point Approach

2018 6th RSI International Conference on Robotics and Mechatronics (IcRoM)(2018)

引用 0|浏览2
暂无评分
摘要
The main target of this paper is to present a Push Recovery (PR) controller for the NAO H21 robot using Model Predictive Control (MPC) method. In order to compensate the applied push the hip strategy, the ankle strategy, the stepping strategy, and the upper-limb rotation strategy have been used. For balance recovery, the Capture Point (CP) position should be guided to the preferred position. For this purpose, the modulation of the Zero Moment Point (ZMP) position and the Centroidal Moment Pivot (CMP) position should be done. Hence, the control of the CP is done using the MPC. Two situations are considered: until the CP position is out of the support polygon the CMP modulation is used and until the CP position is into it the ZMP modulating is used. This strategy is executed on the NAO H21 in dealing with different pushes successfully.
更多
查看译文
关键词
Humanoid robots,Mathematical model,Hip,Collision avoidance,Torque,Cost function
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要