Adaptive Gait Acquisition through Learning Dynamic Stimulus Instinct of Bipedal Robot

Zhang, Yuanxi; Chen, Xuechao; Meng, Fei; Yu, Zhangguo; Du, Yidong; Zhou, Zishun; Gao, Junyao

doi:10.3390/biomimetics9060310

Open AccessArticle

Adaptive Gait Acquisition through Learning Dynamic Stimulus Instinct of Bipedal Robot

by

Yuanxi Zhang

¹,

Xuechao Chen

^1,2

,

Fei Meng

^1,2,*

,

Zhangguo Yu

^1,2

,

Yidong Du

¹,

Zishun Zhou

¹

and

Junyao Gao

^1,2

¹

School of Mechatronical Engineering, Beijing Institute of Technology, Beijing 100081, China

²

Key Laboratory of Biomimetic Robots and Systems, Ministry of Education, Beijing 100081, China

^*

Author to whom correspondence should be addressed.

Biomimetics 2024, 9(6), 310; https://doi.org/10.3390/biomimetics9060310

Submission received: 16 April 2024 / Revised: 11 May 2024 / Accepted: 16 May 2024 / Published: 22 May 2024

(This article belongs to the Special Issue Recent Advances in Robotics and Biomimetics)

Download

Browse Figures

Versions Notes

Abstract

Standard alternating leg motions serve as the foundation for simple bipedal gaits, and the effectiveness of the fixed stimulus signal has been proved in recent studies. However, in order to address perturbations and imbalances, robots require more dynamic gaits. In this paper, we introduce dynamic stimulus signals together with a bipedal locomotion policy into reinforcement learning (RL). Through the learned stimulus frequency policy, we induce the bipedal robot to obtain both three-dimensional (3D) locomotion and an adaptive gait under disturbance without relying on an explicit and model-based gait in both the training stage and deployment. In addition, a set of specialized reward functions focusing on reliable frequency reflections is used in our framework to ensure correspondence between locomotion features and the dynamic stimulus. Moreover, we demonstrate efficient sim-to-real transfer, making a bipedal robot called BITeno achieve robust locomotion and disturbance resistance, even in extreme situations of foot sliding in the real world. In detail, under a sudden change in torso velocity of

- 1.2

m/s in 0.65 s, the recovery time is within 1.5–2.0 s.

Keywords: reinforcement learning; bipedal robot; adaptive locomotion; period dynamic gait

Share and Cite

MDPI and ACS Style

Zhang, Y.; Chen, X.; Meng, F.; Yu, Z.; Du, Y.; Zhou, Z.; Gao, J. Adaptive Gait Acquisition through Learning Dynamic Stimulus Instinct of Bipedal Robot. Biomimetics 2024, 9, 310. https://doi.org/10.3390/biomimetics9060310

AMA Style

Zhang Y, Chen X, Meng F, Yu Z, Du Y, Zhou Z, Gao J. Adaptive Gait Acquisition through Learning Dynamic Stimulus Instinct of Bipedal Robot. Biomimetics. 2024; 9(6):310. https://doi.org/10.3390/biomimetics9060310

Chicago/Turabian Style

Zhang, Yuanxi, Xuechao Chen, Fei Meng, Zhangguo Yu, Yidong Du, Zishun Zhou, and Junyao Gao. 2024. "Adaptive Gait Acquisition through Learning Dynamic Stimulus Instinct of Bipedal Robot" Biomimetics 9, no. 6: 310. https://doi.org/10.3390/biomimetics9060310

APA Style

Zhang, Y., Chen, X., Meng, F., Yu, Z., Du, Y., Zhou, Z., & Gao, J. (2024). Adaptive Gait Acquisition through Learning Dynamic Stimulus Instinct of Bipedal Robot. Biomimetics, 9(6), 310. https://doi.org/10.3390/biomimetics9060310

Article Menu

Adaptive Gait Acquisition through Learning Dynamic Stimulus Instinct of Bipedal Robot

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI