Formation Control of Multi-quadrotors Based on Deep Q-learning

Roujin Mousavifard; Khalil Alipour; Mohammad Amin Najafqolian; Payam Zarafshan

doi:10.1109/ICRoM57054.2022.10025251

Formation Control of Multi-quadrotors Based on Deep Q-learning

Roujin Mousavifard, Khalil Alipour, Mohammad Amin Najafqolian, Payam Zarafshan

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

5 Citations (Scopus)

Abstract

The purpose of this study is to address a model-free formation problem for a team of quadrotors. A cascade controller including a tracking controller and an attitude controller, is developed. The assumptions preserve the nonlinearity and the under-actuation of the model. The tracking controller uses reinforcement learning to develop a model-free online controller. Moreover, the attitude controller is equipped with an actor-critic neural network to solve the nonlinearity issue. The whole formation leads with a virtual leader in the center of the predesigned formation. Simulation results of multiaerial vehicles including four heterogeneous quadrotors, demonstrate the effectiveness of the proposed controller.

Original language	English
Title of host publication	10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	172-177
Number of pages	6
ISBN (Electronic)	9781665454520
DOIs	https://doi.org/10.1109/ICRoM57054.2022.10025251
Publication status	Published - 2022
Externally published	Yes
Event	10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022 - Tehran, Iran, Islamic Republic of Duration: 15 Nov 2022 → 18 Nov 2022

Publication series

Name	10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022

Conference

Conference	10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022
Country/Territory	Iran, Islamic Republic of
City	Tehran
Period	15/11/22 → 18/11/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

Aerial Vehicles
Cascade Control
Deep Q-learning
Model-free
Tracking Control

Access to Document

10.1109/ICRoM57054.2022.10025251

Cite this

Mousavifard, R., Alipour, K., Najafqolian, M. A., & Zarafshan, P. (2022). Formation Control of Multi-quadrotors Based on Deep Q-learning. In 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022 (pp. 172-177). (10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICRoM57054.2022.10025251

@inproceedings{cf519cb7aca349f4ac5346d7b4a324f1,

title = "Formation Control of Multi-quadrotors Based on Deep Q-learning",

abstract = "The purpose of this study is to address a model-free formation problem for a team of quadrotors. A cascade controller including a tracking controller and an attitude controller, is developed. The assumptions preserve the nonlinearity and the under-actuation of the model. The tracking controller uses reinforcement learning to develop a model-free online controller. Moreover, the attitude controller is equipped with an actor-critic neural network to solve the nonlinearity issue. The whole formation leads with a virtual leader in the center of the predesigned formation. Simulation results of multiaerial vehicles including four heterogeneous quadrotors, demonstrate the effectiveness of the proposed controller.",

keywords = "Aerial Vehicles, Cascade Control, Deep Q-learning, Model-free, Tracking Control",

author = "Roujin Mousavifard and Khalil Alipour and Najafqolian, {Mohammad Amin} and Payam Zarafshan",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022 ; Conference date: 15-11-2022 Through 18-11-2022",

year = "2022",

doi = "10.1109/ICRoM57054.2022.10025251",

language = "English",

series = "10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "172--177",

booktitle = "10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022",

}

Mousavifard, R, Alipour, K, Najafqolian, MA & Zarafshan, P 2022, Formation Control of Multi-quadrotors Based on Deep Q-learning. in 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022. 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022, Institute of Electrical and Electronics Engineers Inc., pp. 172-177, 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022, Tehran, Iran, Islamic Republic of, 15/11/22. https://doi.org/10.1109/ICRoM57054.2022.10025251

Formation Control of Multi-quadrotors Based on Deep Q-learning. / Mousavifard, Roujin; Alipour, Khalil; Najafqolian, Mohammad Amin et al.
10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022. Institute of Electrical and Electronics Engineers Inc., 2022. p. 172-177 (10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022).

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

TY - GEN

T1 - Formation Control of Multi-quadrotors Based on Deep Q-learning

AU - Mousavifard, Roujin

AU - Alipour, Khalil

AU - Najafqolian, Mohammad Amin

AU - Zarafshan, Payam

PY - 2022

Y1 - 2022

N2 - The purpose of this study is to address a model-free formation problem for a team of quadrotors. A cascade controller including a tracking controller and an attitude controller, is developed. The assumptions preserve the nonlinearity and the under-actuation of the model. The tracking controller uses reinforcement learning to develop a model-free online controller. Moreover, the attitude controller is equipped with an actor-critic neural network to solve the nonlinearity issue. The whole formation leads with a virtual leader in the center of the predesigned formation. Simulation results of multiaerial vehicles including four heterogeneous quadrotors, demonstrate the effectiveness of the proposed controller.

AB - The purpose of this study is to address a model-free formation problem for a team of quadrotors. A cascade controller including a tracking controller and an attitude controller, is developed. The assumptions preserve the nonlinearity and the under-actuation of the model. The tracking controller uses reinforcement learning to develop a model-free online controller. Moreover, the attitude controller is equipped with an actor-critic neural network to solve the nonlinearity issue. The whole formation leads with a virtual leader in the center of the predesigned formation. Simulation results of multiaerial vehicles including four heterogeneous quadrotors, demonstrate the effectiveness of the proposed controller.

KW - Aerial Vehicles

KW - Cascade Control

KW - Deep Q-learning

KW - Model-free

KW - Tracking Control

UR - http://www.scopus.com/inward/record.url?scp=85148055852&partnerID=8YFLogxK

U2 - 10.1109/ICRoM57054.2022.10025251

DO - 10.1109/ICRoM57054.2022.10025251

M3 - Conference Paper

AN - SCOPUS:85148055852

T3 - 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022

SP - 172

EP - 177

BT - 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022

Y2 - 15 November 2022 through 18 November 2022

ER -

Mousavifard R, Alipour K, Najafqolian MA, Zarafshan P. Formation Control of Multi-quadrotors Based on Deep Q-learning. In 10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 172-177. (10th RSI International Conference on Robotics and Mechatronics, ICRoM 2022). doi: 10.1109/ICRoM57054.2022.10025251

Formation Control of Multi-quadrotors Based on Deep Q-learning

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this