Learning-Based Propulsion Control for Amphibious Quadruped Robots With Dynamic Adaptation to Changing Environment

Qingfeng Yao; Linghan Meng; Qifeng Zhang; Jing Zhao; Joni Pajarinen; Xiaohui Wang; Zhibin Li; Cong Wang

doi:10.1109/LRA.2023.3323893

Learning-Based Propulsion Control for Amphibious Quadruped Robots With Dynamic Adaptation to Changing Environment

Qingfeng Yao, Linghan Meng, Qifeng Zhang, Jing Zhao, Joni Pajarinen, Xiaohui Wang, Zhibin Li, Cong Wang^*

^*Corresponding author for this work

Research output: Contribution to journal › Article › Scientific › peer-review

9 Citations (Scopus)

55 Downloads (Pure)

Abstract

This letter proposes a learning-based adaptive propulsion control (APC) method for a quadruped robot integrated with thrusters in amphibious environments, allowing it to move efficiently in water while maintaining its ground locomotion capabilities. We designed the specific reinforcement learning method to train the neural network to perform the vector propulsion control. Our approach coordinates the legs and propeller, enabling the robot to achieve speed and trajectory tracking tasks in the presence of actuator failures and unknown disturbances. Our simulated validations of the robot in water demonstrate the effectiveness of the trained neural network to predict the disturbances and actuator failures based on historical information, showing that the framework is adaptable to changing environments and is suitable for use in dynamically changing situations. Our proposed approach is suited to the hardware augmentation of quadruped robots to create avenues in the field of amphibious robotics and expand the use of quadruped robots in various applications.

Original language	English
Pages (from-to)	7889-7896
Number of pages	8
Journal	IEEE Robotics and Automation Letters
Volume	8
Issue number	12
DOIs	https://doi.org/10.1109/LRA.2023.3323893
Publication status	Published - 1 Dec 2023
MoE publication type	A1 Journal article-refereed

Funding

This work was supported in part by the Applied Basic Research Program of Liaoning Province under Grant 2022020403-JH2/1013, in part by the National Key Research and Development Program of China under Grant 2022YFB4701900, in part by the National Natural Science Foundation of China under Grant 61821005, in part by the Open Research Project of the State Key Laboratory of Industrial Control Technology, Zhejiang University, China under Grant ICT2023B50, in part by the Program of China Scholarship Council, and in part by the JIANG Xinsong Innovation Fund under Grant E2510202.

Keywords

amphibious robots
Quadruped robots
reinforcement learning
robot learning

Access to Document

10.1109/LRA.2023.3323893

Learning-Based Propulsion Control for Amphibious Quadruped RobotsAccepted author manuscript, 5.67 MB

Cite this

@article{1dad93ca09db48929229bbfd7572a9de,

title = "Learning-Based Propulsion Control for Amphibious Quadruped Robots With Dynamic Adaptation to Changing Environment",

abstract = "This letter proposes a learning-based adaptive propulsion control (APC) method for a quadruped robot integrated with thrusters in amphibious environments, allowing it to move efficiently in water while maintaining its ground locomotion capabilities. We designed the specific reinforcement learning method to train the neural network to perform the vector propulsion control. Our approach coordinates the legs and propeller, enabling the robot to achieve speed and trajectory tracking tasks in the presence of actuator failures and unknown disturbances. Our simulated validations of the robot in water demonstrate the effectiveness of the trained neural network to predict the disturbances and actuator failures based on historical information, showing that the framework is adaptable to changing environments and is suitable for use in dynamically changing situations. Our proposed approach is suited to the hardware augmentation of quadruped robots to create avenues in the field of amphibious robotics and expand the use of quadruped robots in various applications.",

keywords = "amphibious robots, Quadruped robots, reinforcement learning, robot learning",

author = "Qingfeng Yao and Linghan Meng and Qifeng Zhang and Jing Zhao and Joni Pajarinen and Xiaohui Wang and Zhibin Li and Cong Wang",

note = "Funding Information: This work was supported in part by the Applied Basic Research Program of Liaoning Province under Grant 2022020403-JH2/1013, in part by the National Key Research and Development Program of China under Grant 2022YFB4701900, in part by the National Natural Science Foundation of China under Grant 61821005, in part by the Open Research Project of the State Key Laboratory of Industrial Control Technology, Zhejiang University, China under Grant ICT2023B50, in part by the Program of China Scholarship Council, and in part by the JIANG Xinsong Innovation Fund under Grant E2510202. Publisher Copyright: {\textcopyright} 2016 IEEE.",

year = "2023",

month = dec,

day = "1",

doi = "10.1109/LRA.2023.3323893",

language = "English",

volume = "8",

pages = "7889--7896",

journal = "IEEE Robotics and Automation Letters",

issn = "2377-3766",

publisher = "IEEE",

number = "12",

}

TY - JOUR

T1 - Learning-Based Propulsion Control for Amphibious Quadruped Robots With Dynamic Adaptation to Changing Environment

AU - Yao, Qingfeng

AU - Meng, Linghan

AU - Zhang, Qifeng

AU - Zhao, Jing

AU - Pajarinen, Joni

AU - Wang, Xiaohui

AU - Li, Zhibin

AU - Wang, Cong

N1 - Funding Information: This work was supported in part by the Applied Basic Research Program of Liaoning Province under Grant 2022020403-JH2/1013, in part by the National Key Research and Development Program of China under Grant 2022YFB4701900, in part by the National Natural Science Foundation of China under Grant 61821005, in part by the Open Research Project of the State Key Laboratory of Industrial Control Technology, Zhejiang University, China under Grant ICT2023B50, in part by the Program of China Scholarship Council, and in part by the JIANG Xinsong Innovation Fund under Grant E2510202. Publisher Copyright: © 2016 IEEE.

PY - 2023/12/1

Y1 - 2023/12/1

N2 - This letter proposes a learning-based adaptive propulsion control (APC) method for a quadruped robot integrated with thrusters in amphibious environments, allowing it to move efficiently in water while maintaining its ground locomotion capabilities. We designed the specific reinforcement learning method to train the neural network to perform the vector propulsion control. Our approach coordinates the legs and propeller, enabling the robot to achieve speed and trajectory tracking tasks in the presence of actuator failures and unknown disturbances. Our simulated validations of the robot in water demonstrate the effectiveness of the trained neural network to predict the disturbances and actuator failures based on historical information, showing that the framework is adaptable to changing environments and is suitable for use in dynamically changing situations. Our proposed approach is suited to the hardware augmentation of quadruped robots to create avenues in the field of amphibious robotics and expand the use of quadruped robots in various applications.

AB - This letter proposes a learning-based adaptive propulsion control (APC) method for a quadruped robot integrated with thrusters in amphibious environments, allowing it to move efficiently in water while maintaining its ground locomotion capabilities. We designed the specific reinforcement learning method to train the neural network to perform the vector propulsion control. Our approach coordinates the legs and propeller, enabling the robot to achieve speed and trajectory tracking tasks in the presence of actuator failures and unknown disturbances. Our simulated validations of the robot in water demonstrate the effectiveness of the trained neural network to predict the disturbances and actuator failures based on historical information, showing that the framework is adaptable to changing environments and is suitable for use in dynamically changing situations. Our proposed approach is suited to the hardware augmentation of quadruped robots to create avenues in the field of amphibious robotics and expand the use of quadruped robots in various applications.

KW - amphibious robots

KW - Quadruped robots

KW - reinforcement learning

KW - robot learning

UR - https://www.scopus.com/pages/publications/85174822124

U2 - 10.1109/LRA.2023.3323893

DO - 10.1109/LRA.2023.3323893

M3 - Article

AN - SCOPUS:85174822124

SN - 2377-3766

VL - 8

SP - 7889

EP - 7896

JO - IEEE Robotics and Automation Letters

JF - IEEE Robotics and Automation Letters

IS - 12

ER -

Learning-Based Propulsion Control for Amphibious Quadruped Robots With Dynamic Adaptation to Changing Environment

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this