Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning

Ming-Chieh Yang; Hooman Samani; Kening Zhu

doi:10.1007/978-3-030-26118-4_31

Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning

Ming-Chieh Yang, Hooman Samani^*, Kening Zhu

^*Corresponding author for this work

School of Creative Media

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

4 Citations (Scopus)

Abstract

The locomotion of legged robot is often controlled by predefined gaits, and this approach works well when all joints and motors are operating normally. However, walking legged robots usually have high risk of being damaged during operation, causing the breakdown of the robotic joints. In this paper, we introduce a reinforcement learning based approach for the legged robot to generate real-time locomotion response to the emergence of locomotion breakdown. Our approach detects the functionality of the available joints, substitutes the pre-defined gaits with proper gait function accordingly, and upgrades the gait-generation function by Q-Learning for the proper locomotion.

Original language	English
Title of host publication	Interactive Collaborative Robotics
Subtitle of host publication	Proceedings
Editors	Andrey Ronzhin, Gerhard Rigoll, Roman Meshcheryakov
Publisher	Springer Nature Switzerland AG
Pages	320-329
ISBN (Electronic)	9783030261184
ISBN (Print)	9783030261177
DOIs	https://doi.org/10.1007/978-3-030-26118-4_31
Publication status	Published - Aug 2019
Event	4th International Conference on Interactive Collaborative Robotics (ICR 2019) - Istanbul, Türkiye Duration: 20 Aug 2019 → 25 Aug 2019

Publication series

Name	Lecture Notes in Artificial Intelligence
Volume	11659 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	4th International Conference on Interactive Collaborative Robotics (ICR 2019)
Abbreviated title	ICR 2019
Place	Türkiye
City	Istanbul
Period	20/08/19 → 25/08/19

Research Keywords

Emergency response
Hexapod robot
Q-Learning
Reinforcement learning

Access Output

10.1007/978-3-030-26118-4_31

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Yang, M.-C., Samani, H., & Zhu, K. (2019). Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning. In A. Ronzhin, G. Rigoll, & R. Meshcheryakov (Eds.), Interactive Collaborative Robotics: Proceedings (pp. 320-329). (Lecture Notes in Artificial Intelligence; Vol. 11659 LNAI). Springer Nature Switzerland AG. https://doi.org/10.1007/978-3-030-26118-4_31

@inproceedings{e2d26b65e3944bb4926cd992a339f73d,

title = "Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning",

abstract = "The locomotion of legged robot is often controlled by predefined gaits, and this approach works well when all joints and motors are operating normally. However, walking legged robots usually have high risk of being damaged during operation, causing the breakdown of the robotic joints. In this paper, we introduce a reinforcement learning based approach for the legged robot to generate real-time locomotion response to the emergence of locomotion breakdown. Our approach detects the functionality of the available joints, substitutes the pre-defined gaits with proper gait function accordingly, and upgrades the gait-generation function by Q-Learning for the proper locomotion.",

keywords = "Emergency response, Hexapod robot, Q-Learning, Reinforcement learning",

author = "Ming-Chieh Yang and Hooman Samani and Kening Zhu",

year = "2019",

month = aug,

doi = "10.1007/978-3-030-26118-4_31",

language = "English",

isbn = "9783030261177",

series = "Lecture Notes in Artificial Intelligence",

publisher = "Springer Nature Switzerland AG",

pages = "320--329",

editor = "Andrey Ronzhin and Gerhard Rigoll and Roman Meshcheryakov",

booktitle = "Interactive Collaborative Robotics",

address = "Switzerland",

note = "4th International Conference on Interactive Collaborative Robotics (ICR 2019), ICR 2019 ; Conference date: 20-08-2019 Through 25-08-2019",

}

Yang, M-C, Samani, H & Zhu, K 2019, Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning. in A Ronzhin, G Rigoll & R Meshcheryakov (eds), Interactive Collaborative Robotics: Proceedings. Lecture Notes in Artificial Intelligence, vol. 11659 LNAI, Springer Nature Switzerland AG, pp. 320-329, 4th International Conference on Interactive Collaborative Robotics (ICR 2019), Istanbul, Türkiye, 20/08/19. https://doi.org/10.1007/978-3-030-26118-4_31

Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning. / Yang, Ming-Chieh; Samani, Hooman; Zhu, Kening.
Interactive Collaborative Robotics: Proceedings. ed. / Andrey Ronzhin; Gerhard Rigoll; Roman Meshcheryakov. Springer Nature Switzerland AG, 2019. p. 320-329 (Lecture Notes in Artificial Intelligence; Vol. 11659 LNAI).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning

AU - Yang, Ming-Chieh

AU - Samani, Hooman

AU - Zhu, Kening

PY - 2019/8

Y1 - 2019/8

N2 - The locomotion of legged robot is often controlled by predefined gaits, and this approach works well when all joints and motors are operating normally. However, walking legged robots usually have high risk of being damaged during operation, causing the breakdown of the robotic joints. In this paper, we introduce a reinforcement learning based approach for the legged robot to generate real-time locomotion response to the emergence of locomotion breakdown. Our approach detects the functionality of the available joints, substitutes the pre-defined gaits with proper gait function accordingly, and upgrades the gait-generation function by Q-Learning for the proper locomotion.

AB - The locomotion of legged robot is often controlled by predefined gaits, and this approach works well when all joints and motors are operating normally. However, walking legged robots usually have high risk of being damaged during operation, causing the breakdown of the robotic joints. In this paper, we introduce a reinforcement learning based approach for the legged robot to generate real-time locomotion response to the emergence of locomotion breakdown. Our approach detects the functionality of the available joints, substitutes the pre-defined gaits with proper gait function accordingly, and upgrades the gait-generation function by Q-Learning for the proper locomotion.

KW - Emergency response

KW - Hexapod robot

KW - Q-Learning

KW - Reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=85071452335&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85071452335&origin=recordpage

U2 - 10.1007/978-3-030-26118-4_31

DO - 10.1007/978-3-030-26118-4_31

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9783030261177

T3 - Lecture Notes in Artificial Intelligence

SP - 320

EP - 329

BT - Interactive Collaborative Robotics

A2 - Ronzhin, Andrey

A2 - Rigoll, Gerhard

A2 - Meshcheryakov, Roman

PB - Springer Nature Switzerland AG

T2 - 4th International Conference on Interactive Collaborative Robotics (ICR 2019)

Y2 - 20 August 2019 through 25 August 2019

ER -

Yang MC, Samani H, Zhu K. Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning. In Ronzhin A, Rigoll G, Meshcheryakov R, editors, Interactive Collaborative Robotics: Proceedings. Springer Nature Switzerland AG. 2019. p. 320-329. (Lecture Notes in Artificial Intelligence). Epub 2019 Aug 13. doi: 10.1007/978-3-030-26118-4_31

Emergency-response locomotion of hexapod robot with heuristic reinforcement learning using Q-learning

Abstract

Publication series

Conference

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this