Dynamic Task Allocation for Robotic Edge System Resilience Using Deep Reinforcement Learning
IEEE Transactions on Systems, Man, and Cybernetics: Systems · 2024-03 · 54 (3) · 1438-1450
DOI: 10.1109/tsmc.2023.3327959 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2024-03. Verfasst von Mahbuba Afrin und 5 weiteren Autor:innen. Die Arbeit wurde bisher 25-mal zitiert (Crossref) und nennt 27 Literaturangaben. Gefördert von: Australian Research Council Linkage Project, Data61, CSIRO.
Abstract (Originalsprache)
Incorporating edge and cloud computing with robotics provides extended options for robots to perform real-time sensing and actuation operations in various cyber–physical systems (CPSs), including smart farms. Such systems are prone to uncertain failures triggered by mechanical disruptions. Consequently, the overall system performance degrades, primarily when location-specific tasks are already assigned to a faulty robot and require immediate recovery. Using edge and cloud computing resources is not always feasible due to communication and latency constraints. Therefore, this article exclusively focuses on harnessing the mobility of robots to support the computation tasks affected by uncertain failures of previously assigned robots and ensure faster resiliency management by relocating active robots near task sources. The proposed mobility-as-a-resilience-service (MaaRS) is formulated using a Markov decision process (MDP). Later, an edge server proximal to the robots is trained using deep reinforcement learning (DRL) to assign tasks among the robots. Specifically, a multiple deep $Q$ -network (MDQN)-based dynamic task allocation mechanism is proposed to converge to a solution exploring reward uncertainties with the best exploitation. Numerical evaluation using Python and TensorFlow validates the effectiveness of the proposed approach compared to other benchmarks.
Förderung
- Australian Research Council Linkage Project
- Data61, CSIRO
Zitieren
APA
Afrin, M., Jin, J., Rahman, A., Li, S., Tian, Y., & Li, Y. (2024). Dynamic Task Allocation for Robotic Edge System Resilience Using Deep Reinforcement Learning. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 54(3), 1438-1450. https://doi.org/10.1109/tsmc.2023.3327959
Vancouver
Afrin M, Jin J, Rahman A, Li S, Tian Y, Li Y. Dynamic Task Allocation for Robotic Edge System Resilience Using Deep Reinforcement Learning. IEEE Transactions on Systems, Man, and Cybernetics: Systems. 2024;54(3):1438-1450. doi:10.1109/tsmc.2023.3327959
BibTeX
@article{Afrin2024,
author = {Afrin, Mahbuba and Jin, Jiong and Rahman, Ashfaqur and Li, Shi and Tian, Yu-Chu and Li, Yan},
title = {Dynamic Task Allocation for Robotic Edge System Resilience Using Deep Reinforcement Learning},
journal = {IEEE Transactions on Systems, Man, and Cybernetics: Systems},
year = {2024},
volume = {54},
number = {3},
pages = {1438--1450},
doi = {10.1109/tsmc.2023.3327959}
}
EUVIMED