logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Data-driven iterative learning heading control for USVs under aperiodic DoS attacks and sensor saturation

delete2026-08-21
delete0
PRE
AI
B
Bin Hu *
X
Xiu-Ying Huang
D
De-zheng Zeng
DOI：10.1016/j.oceaneng.2026.127642delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• In contrast to model-based approaches, the heading control problem for USVs is addressed within a data-driven framework, thereby eliminating the requirement for an accurate mathematical model. • Compared with existing MFAILC and MFAC scheme designed for USVs, the proposed DDILC approach simultaneously handling aperiodic DoS attacks and sensor saturation. Rigorous asymptotic convergence is guaranteed with a tunable factor that overcomes CFDL method inapplicability, while precise heading angle tracking and angular velocity synchronization are simultaneously achieved. • A compensation strategy is proposed to mitigate data losses induced by aperiodic DoS attacks. The proposed compensation strategy reduces the averaged root mean square error from 0.118 to 0.071 after 100 iterations under DoS attacks. Abstract This paper proposes a data-driven iterative learning control (DDILC) scheme for unmanned surface vehicles (USVs) subject to aperiodic Denial-of-Service (DoS) attacks and sensor saturation. To overcome the inapplicability of the compact form dynamic linearization (CFDL) method caused by the distinct dynamics of the heading control subsystem, a modified output formulation incorporating a tuning factor is introduced. Subsequently, a dynamic linearization technique is applied along the iteration axis to transform the nonlinear USV subsystem into an equivalent data-driven model. A compensation mechanism is constructed to counteract the detrimental effects of lost or corrupted signals caused by DoS attacks. Based solely on the available control inputs and saturated output measurements, a DDILC algorithm is synthesized to guarantee that the USV follows a prescribed trajectory. A rigorous convergence analysis demonstrates that the tracking error converges to zero asymptotically under the proposed scheme. Finally, Simulation results are provided to confirm the effectiveness of the developed heading control strategy.
Keywords:
Data-driven iterative learning control (DDILC)
Unmanned surface vehicles (USVs)
Denial of service (DoS) attacks
Compact form dynamic linearization (CFDL)
Sensor saturation

Journal

Ocean Engineering cover
Ocean Engineering
IF:
5.5
Papers:
5.8K
Citations:
7.6W

Organization

No organization information available