Swing Foot Pose Control Disturbance Overcoming Algorithm Based on Reference ZMP Preview Controller for Improving Humanoid Walking Stability

被引：0

作者：

Chung, Eunsoo ^{[1
]}

Jung, Haewon ^{[2
]}

Chun, Yeonghun ^{[1
]}

Lee, Wonyong ^{[2
]}

Lee, Jangwon ^{[2
]}

Ahn, Minsung ^{[3
]}

Han, Jeakweon ^{[4
]}

机构：

[1] Hanyang Univ, Dept Convergence Robot Syst, Ansan 15588, South Korea

[2] Hanyang Univ, Dept Robot, Ansan 15588, South Korea

[3] Univ Calif Los Angeles, Mech & Aerosp Engn, Los Angeles, CA 90095 USA

[4] Hanyang Univ, Dept Robot, Ansan 15588, South Korea

来源：

ROBOCUP 2023: ROBOT WORLD CUP XXVI | 2024年 / 14140卷

关键词：

Overcoming Disturbance; Humanoid Robot; Preview Controller; Bipedal Walking; BIPED ROBOT WALKING; OPTIMIZATION;

D O I：

10.1007/978-3-031-55015-7_16

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

To improve the walking stability of humanoid robots, it is necessary to increase the performance of overcoming external disturbance. In this paper, step modification strategy based on the predicted Zero Moment Point (ZMP) is applied to the Reference ZMP Preview Controller to increase the overcoming disturbance performance of humanoid robots. It also uses a Horizon Swing Foot strategy that can keep the ankle of the Swing phase parallel to the ground to increase stability when changing the Step. The proposed algorithm was verified through experiments and confirmed that the Horizon Swing Foot strategy can overcome up to 14.85 kg center dot m/s of impact. By applying the proposed algorithm to humanoid robots, it is expected to increase bipedal walking stability and prevent damage and safety accidents caused by the robot falling.

引用

页码：191 / 202

页数：12