Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning

Zihan Xie; Yongyuan Li; Yingzi Guan; Changzhu Wei

doi:10.1007/978-981-19-3998-3_152

Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning

Zihan Xie
, Yongyuan Li
, Yingzi Guan^*
, Changzhu Wei

^*Corresponding author for this work

School of Astronautics, Harbin Institute of Technology
China Aerospace Science and Technology Corporation

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

3 Scopus citations

Abstract

In this paper, a sliding mode controller with fixed time convergence is designed for the UAV formation control problem, and a controller parameter online tuning method is proposed as well. First, the formation control problem is described and a formation control model is developed. Second, a formation controller based on terminal sliding mode theory is designed, whose fixed-time convergence is proved theoretically. Besides, aiming to the problem that sliding mode controller parameters are difficult to determine and lack of adaptability, a controller parameter tuning method based on actor-critic framework is proposed by taking maximum acceleration and energy cost into account. Simulation results show that the control method proposed in this paper can generate the desired formation quickly, and effectively reduce the convergence time and energy cost, with good dynamic performance.

Original language	English
Title of host publication	Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control
Editors	Zhang Ren, Yongzhao Hua, Mengyi Wang
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	1628-1639
Number of pages	12
ISBN (Print)	9789811939976
DOIs	https://doi.org/10.1007/978-981-19-3998-3_152
State	Published - 2023
Externally published	Yes
Event	5th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2021 - Shenzhen, China Duration: 19 Jan 2022 → 22 Jan 2022

Publication series

Name	Lecture Notes in Electrical Engineering
Volume	934 LNEE
ISSN (Print)	1876-1100
ISSN (Electronic)	1876-1119

Conference

Conference	5th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2021
Country/Territory	China
City	Shenzhen
Period	19/01/22 → 22/01/22

Keywords

Formation control
Reinforcement learning
Sliding mode

Access to Document

10.1007/978-981-19-3998-3_152

Cite this

Xie, Z., Li, Y., Guan, Y., & Wei, C. (2023). Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning. In Z. Ren, Y. Hua, & M. Wang (Eds.), Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control (pp. 1628-1639). (Lecture Notes in Electrical Engineering; Vol. 934 LNEE). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-19-3998-3_152

Xie, Zihan ; Li, Yongyuan ; Guan, Yingzi et al. / Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning. Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control. editor / Zhang Ren ; Yongzhao Hua ; Mengyi Wang. Springer Science and Business Media Deutschland GmbH, 2023. pp. 1628-1639 (Lecture Notes in Electrical Engineering).

@inproceedings{518787f601174a3ebd8ac59e9f316b33,

title = "Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning",

abstract = "In this paper, a sliding mode controller with fixed time convergence is designed for the UAV formation control problem, and a controller parameter online tuning method is proposed as well. First, the formation control problem is described and a formation control model is developed. Second, a formation controller based on terminal sliding mode theory is designed, whose fixed-time convergence is proved theoretically. Besides, aiming to the problem that sliding mode controller parameters are difficult to determine and lack of adaptability, a controller parameter tuning method based on actor-critic framework is proposed by taking maximum acceleration and energy cost into account. Simulation results show that the control method proposed in this paper can generate the desired formation quickly, and effectively reduce the convergence time and energy cost, with good dynamic performance.",

keywords = "Formation control, Reinforcement learning, Sliding mode",

author = "Zihan Xie and Yongyuan Li and Yingzi Guan and Changzhu Wei",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.; 5th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2021 ; Conference date: 19-01-2022 Through 22-01-2022",

year = "2023",

doi = "10.1007/978-981-19-3998-3\_152",

language = "英语",

isbn = "9789811939976",

series = "Lecture Notes in Electrical Engineering",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "1628--1639",

editor = "Zhang Ren and Yongzhao Hua and Mengyi Wang",

booktitle = "Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control",

address = "德国",

}

Xie, Z, Li, Y, Guan, Y & Wei, C 2023, Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning. in Z Ren, Y Hua & M Wang (eds), Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control. Lecture Notes in Electrical Engineering, vol. 934 LNEE, Springer Science and Business Media Deutschland GmbH, pp. 1628-1639, 5th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2021, Shenzhen, China, 19/01/22. https://doi.org/10.1007/978-981-19-3998-3_152

Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning. / Xie, Zihan; Li, Yongyuan; Guan, Yingzi et al.
Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control. ed. / Zhang Ren; Yongzhao Hua; Mengyi Wang. Springer Science and Business Media Deutschland GmbH, 2023. p. 1628-1639 (Lecture Notes in Electrical Engineering; Vol. 934 LNEE).

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

TY - GEN

T1 - Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning

AU - Xie, Zihan

AU - Li, Yongyuan

AU - Guan, Yingzi

AU - Wei, Changzhu

PY - 2023

Y1 - 2023

N2 - In this paper, a sliding mode controller with fixed time convergence is designed for the UAV formation control problem, and a controller parameter online tuning method is proposed as well. First, the formation control problem is described and a formation control model is developed. Second, a formation controller based on terminal sliding mode theory is designed, whose fixed-time convergence is proved theoretically. Besides, aiming to the problem that sliding mode controller parameters are difficult to determine and lack of adaptability, a controller parameter tuning method based on actor-critic framework is proposed by taking maximum acceleration and energy cost into account. Simulation results show that the control method proposed in this paper can generate the desired formation quickly, and effectively reduce the convergence time and energy cost, with good dynamic performance.

AB - In this paper, a sliding mode controller with fixed time convergence is designed for the UAV formation control problem, and a controller parameter online tuning method is proposed as well. First, the formation control problem is described and a formation control model is developed. Second, a formation controller based on terminal sliding mode theory is designed, whose fixed-time convergence is proved theoretically. Besides, aiming to the problem that sliding mode controller parameters are difficult to determine and lack of adaptability, a controller parameter tuning method based on actor-critic framework is proposed by taking maximum acceleration and energy cost into account. Simulation results show that the control method proposed in this paper can generate the desired formation quickly, and effectively reduce the convergence time and energy cost, with good dynamic performance.

KW - Formation control

KW - Reinforcement learning

KW - Sliding mode

UR - https://www.scopus.com/pages/publications/85135851078

U2 - 10.1007/978-981-19-3998-3_152

DO - 10.1007/978-981-19-3998-3_152

M3 - 会议稿件

AN - SCOPUS:85135851078

SN - 9789811939976

T3 - Lecture Notes in Electrical Engineering

SP - 1628

EP - 1639

BT - Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control

A2 - Ren, Zhang

A2 - Hua, Yongzhao

A2 - Wang, Mengyi

PB - Springer Science and Business Media Deutschland GmbH

T2 - 5th Chinese Conference on Swarm Intelligence and Cooperative Control, CCSICC 2021

Y2 - 19 January 2022 through 22 January 2022

ER -

Xie Z, Li Y, Guan Y , Wei C. Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning. In Ren Z, Hua Y, Wang M, editors, Proceedings of 2021 5th Chinese Conference on Swarm Intelligence and Cooperative Control. Springer Science and Business Media Deutschland GmbH. 2023. p. 1628-1639. (Lecture Notes in Electrical Engineering). doi: 10.1007/978-981-19-3998-3_152

Multi-UAV Sliding Mode Formation Control Based on Reinforcement Learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this