Comparison of Backstepping with Reinforcement Learning

Huan Yu; Miroslav Krstic

doi:10.1007/978-3-031-19346-0_7

Comparison of Backstepping with Reinforcement Learning

Huan Yu^*, Miroslav Krstic

^*Corresponding author for this work

Research output: Chapter in Book/Conference Proceeding/Report › Book Chapter › peer-review

Abstract

A reinforcement learning controller, with training performed on an ARZ PDE model, is developed and evaluated for the amount of training necessary, the performance during the training process, and the performance ultimately achieved relative to a model-based PDE backstepping controller that is designed in a single shot and provides stability guarantees.

Original language	English
Title of host publication	Systems and Control
Subtitle of host publication	Foundations and Applications
Publisher	Birkhauser
Pages	125-160
Number of pages	36
DOIs	https://doi.org/10.1007/978-3-031-19346-0_7
Publication status	Published - 2022
Externally published	Yes

Publication series

Name	Systems and Control: Foundations and Applications
ISSN (Print)	2324-9749
ISSN (Electronic)	2324-9757

Bibliographical note

Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Access to Document

10.1007/978-3-031-19346-0_7

Cite this

@inbook{debbcca2d91546ff8ca22de161e09e05,

title = "Comparison of Backstepping with Reinforcement Learning",

abstract = "A reinforcement learning controller, with training performed on an ARZ PDE model, is developed and evaluated for the amount of training necessary, the performance during the training process, and the performance ultimately achieved relative to a model-based PDE backstepping controller that is designed in a single shot and provides stability guarantees.",

author = "Huan Yu and Miroslav Krstic",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.",

year = "2022",

doi = "10.1007/978-3-031-19346-0\_7",

language = "English",

series = "Systems and Control: Foundations and Applications",

publisher = "Birkhauser",

pages = "125--160",

booktitle = "Systems and Control",

}

TY - CHAP

T1 - Comparison of Backstepping with Reinforcement Learning

AU - Yu, Huan

AU - Krstic, Miroslav

PY - 2022

Y1 - 2022

N2 - A reinforcement learning controller, with training performed on an ARZ PDE model, is developed and evaluated for the amount of training necessary, the performance during the training process, and the performance ultimately achieved relative to a model-based PDE backstepping controller that is designed in a single shot and provides stability guarantees.

AB - A reinforcement learning controller, with training performed on an ARZ PDE model, is developed and evaluated for the amount of training necessary, the performance during the training process, and the performance ultimately achieved relative to a model-based PDE backstepping controller that is designed in a single shot and provides stability guarantees.

UR - https://openalex.org/W4312370004

UR - https://www.scopus.com/pages/publications/85144730724

U2 - 10.1007/978-3-031-19346-0_7

DO - 10.1007/978-3-031-19346-0_7

M3 - Book Chapter

T3 - Systems and Control: Foundations and Applications

SP - 125

EP - 160

BT - Systems and Control

PB - Birkhauser

ER -

Comparison of Backstepping with Reinforcement Learning

Abstract

Publication series

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this