Home > Research > Publications & Outputs > Distributed Safe Reinforcement Learning for Mul...

Research

Associated organisational units

Links

Text available via DOI:

View graph of relations

Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review

Published

Standard

Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning. / Lu, Yang; Guo, Yaohua; Zhao, Guoxiang et al.
2021 29th Mediterranean Conference on Control and Automation (MED). IEEE, 2021. (2021 29th Mediterranean Conference on Control and Automation (MED); Vol. 2021).

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSN › Conference contribution/Paper › peer-review

Harvard

Lu, Y, Guo, Y, Zhao, G & Zhu, M 2021, Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning. in 2021 29th Mediterranean Conference on Control and Automation (MED). 2021 29th Mediterranean Conference on Control and Automation (MED), vol. 2021, IEEE. https://doi.org/10.1109/MED51440.2021.9480176

APA

Lu, Y., Guo, Y., Zhao, G., & Zhu, M. (2021). Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning. In 2021 29th Mediterranean Conference on Control and Automation (MED) (2021 29th Mediterranean Conference on Control and Automation (MED); Vol. 2021). IEEE. https://doi.org/10.1109/MED51440.2021.9480176

Vancouver

Lu Y, Guo Y, Zhao G, Zhu M. Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning. In 2021 29th Mediterranean Conference on Control and Automation (MED). IEEE. 2021. (2021 29th Mediterranean Conference on Control and Automation (MED)). Epub 2021 Jun 25. doi: 10.1109/MED51440.2021.9480176

Author

Lu, Yang ; Guo, Yaohua ; Zhao, Guoxiang et al. / Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning. 2021 29th Mediterranean Conference on Control and Automation (MED). IEEE, 2021. (2021 29th Mediterranean Conference on Control and Automation (MED)).

Bibtex

@inproceedings{e8a7a3982b3540c594a55f2c5ab2fbba,
title = "Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning",
abstract = "This paper studies optimal motion planning of multiple mobile robots with collision avoidance. We develop a distributed reinforcement learning algorithm which ensures suboptimal goal reaching and anytime collision avoidance simultaneously. Theoretical results on the convergence of neural network weights, the uniform and ultimate boundedness of system states of the closed-loop system, and anytime collision avoidance are established. Numerical simulations for single integrator and unicycle robots illustrate the effectiveness of our theoretical results.",
author = "Yang Lu and Yaohua Guo and Guoxiang Zhao and Minghui Zhu",
year = "2021",
month = jul,
day = "15",
doi = "10.1109/MED51440.2021.9480176",
language = "English",
isbn = "9781665446600",
series = "2021 29th Mediterranean Conference on Control and Automation (MED)",
publisher = "IEEE",
booktitle = "2021 29th Mediterranean Conference on Control and Automation (MED)",

}

RIS

TY - GEN

T1 - Distributed Safe Reinforcement Learning for Multi-Robot Motion Planning

AU - Lu, Yang

AU - Guo, Yaohua

AU - Zhao, Guoxiang

AU - Zhu, Minghui

PY - 2021/7/15

Y1 - 2021/7/15

N2 - This paper studies optimal motion planning of multiple mobile robots with collision avoidance. We develop a distributed reinforcement learning algorithm which ensures suboptimal goal reaching and anytime collision avoidance simultaneously. Theoretical results on the convergence of neural network weights, the uniform and ultimate boundedness of system states of the closed-loop system, and anytime collision avoidance are established. Numerical simulations for single integrator and unicycle robots illustrate the effectiveness of our theoretical results.

AB - This paper studies optimal motion planning of multiple mobile robots with collision avoidance. We develop a distributed reinforcement learning algorithm which ensures suboptimal goal reaching and anytime collision avoidance simultaneously. Theoretical results on the convergence of neural network weights, the uniform and ultimate boundedness of system states of the closed-loop system, and anytime collision avoidance are established. Numerical simulations for single integrator and unicycle robots illustrate the effectiveness of our theoretical results.

U2 - 10.1109/MED51440.2021.9480176

DO - 10.1109/MED51440.2021.9480176

M3 - Conference contribution/Paper

SN - 9781665446600

T3 - 2021 29th Mediterranean Conference on Control and Automation (MED)

BT - 2021 29th Mediterranean Conference on Control and Automation (MED)

PB - IEEE

ER -