Bias-compensated Q-learning for optimal tracking control under denial-of-service attacks
ISA Transactions · 2026-10 · 177 · 98-108
DOI: 10.1016/j.isatra.2026.06.050 ↗Kurz gefasst
Fachartikel in der Zeitschrift ISA Transactions, 2026-10. Verfasst von Zhenhuan Su und 1 weiteren Autor:innen. Beteiligte Einrichtung: Key Laboratory of IoT and Information Fusion Technology of Zhejiang Province, School of Automation, Hangzhou Dianzi University, 2nd Street 1, Hangzho…. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 49 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Networked control systems are widely adopted in modern applications. Nevertheless, learning optimal control policies from data under communication attacks remains challenging. In this paper, we propose a bias-compensated Q-learning algorithm for knowledge-informed optimal tracking control. The method considers discrete-time linear time-invariant systems subject to control-channel denial-of-service attacks. Through a bias compensation mechanism that addresses control-input losses, the proposed method eliminates the need for explicit expectation computation in Q-function updates and enables direct data-driven learning under DoS attacks. Moreover, the proposed method requires no knowledge of system dynamics beyond the known system structure and supports data collection under arbitrary behavior policies. This design preserves high data efficiency, computational simplicity, and robustness within a model-free framework. Numerical simulations on a planar quadrotor position-tracking task show that a Riccati-equivalent optimal policy can still be learned from data under 40% DoS attacks.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Key Laboratory of IoT and Information Fusion Technology of Zhejiang Province, School of Automation, Hangzhou Dianzi University, 2nd Street 1, Hangzhou, 310018, Zhejiang, China.
- Key Laboratory of IoT and Information Fusion Technology of Zhejiang Province, School of Automation, Hangzhou Dianzi University, 2nd Street 1, Hangzhou, 310018, Zhejiang, China. Electronic address: g.liu@hdu.edu.cn.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Su, Z., & Liu, G. (2026). Bias-compensated Q-learning for optimal tracking control under denial-of-service attacks. ISA Transactions, 177, 98-108. https://doi.org/10.1016/j.isatra.2026.06.050
Vancouver
Su Z, Liu G. Bias-compensated Q-learning for optimal tracking control under denial-of-service attacks. ISA Transactions. 2026;177:98-108. doi:10.1016/j.isatra.2026.06.050
BibTeX
@article{Su2026,
author = {Su, Zhenhuan and Liu, Guangyu},
title = {Bias-compensated Q-learning for optimal tracking control under denial-of-service attacks},
journal = {ISA Transactions},
year = {2026},
volume = {177},
pages = {98--108},
doi = {10.1016/j.isatra.2026.06.050}
}
EUVIMED