EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Robust Finite-Time Neural State Observer-Driven Fault-Tolerant Control of USVs Under Actuator Faults

Wenxue Su, Wei Liu, Yuan Hu, Jingtao Pei, Xingwang Huang

Journal of Marine Science and Engineering · 22.04.2026 · 14 (9) · 766

DOI: 10.3390/jmse14090766 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Marine Science and Engineering, online erschienen am 22.04.2026. Verfasst von Wenxue Su und 4 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shandong Port Technology Project: Research on Ship Autonomous Navigation Perception Control Technology, Construction of Online High-Quality Vocational Education Courses in Shandong Province: Navigation I, Construction of high-quality online courses at the vocational education school level: Navigation Instruments.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To address the actuator fault problem faced by underactuated surface vessels (USVs), this study develops an active fault-tolerant control scheme based on finite-time output feedback. First, a finite-time neural terminal homogeneous state observer with a portional-integral structure is established. High-precision pose reconstruction enables finite-time synchronous reconstruction of unmeasured states. This allows unknown nonlinearities to be explicitly expressed online and incorporated into the compensation channel, significantly reducing the sensitivity of modeling errors to control performance. A neural damping mechanism is used to structurally reconstruct uncertain dynamics and loss-of-effectiveness (LOE) fault factors within the system, thereby constructing an online approximator to achieve real-time identification and compensation of composite uncertainties. This integrates the unknown nonlinearities and fault effects of the original system into an online-updatable estimation channel. Adopting a backstepping-based design methodology, a finite-time hybrid event-triggered control (ETC) architecture is further constructed. By introducing an event-triggered update mechanism at the control layer, the real-time continuous control signal is transformed into a discrete update. Based on Lyapunov stability theory, a comprehensive analysis is carried out to verify the stability of the proposed control scheme. Numerical simulations are finally carried out to validate the effectiveness of the scheme. Simulation results show that the tracking error is reduced by about 93% and 60% compared to the comparison scheme.

Förderung

  • Shandong Port Technology Project: Research on Ship Autonomous Navigation Perception Control Technology
  • Construction of Online High-Quality Vocational Education Courses in Shandong Province: Navigation I
  • Construction of high-quality online courses at the vocational education school level: Navigation Instruments
  • National High-Quality Online Vocational Education Course: Uncovering Navigation

Zitieren

APA

Su, W., Liu, W., Hu, Y., Pei, J., & Huang, X. (2026). Robust Finite-Time Neural State Observer-Driven Fault-Tolerant Control of USVs Under Actuator Faults. Journal of Marine Science and Engineering, 14(9), 766. https://doi.org/10.3390/jmse14090766

Vancouver

Su W, Liu W, Hu Y, Pei J, Huang X. Robust Finite-Time Neural State Observer-Driven Fault-Tolerant Control of USVs Under Actuator Faults. Journal of Marine Science and Engineering. 2026;14(9):766. doi:10.3390/jmse14090766

BibTeX
@article{Su2026,
  author = {Su, Wenxue and Liu, Wei and Hu, Yuan and Pei, Jingtao and Huang, Xingwang},
  title = {Robust Finite-Time Neural State Observer-Driven Fault-Tolerant Control of USVs Under Actuator Faults},
  journal = {Journal of Marine Science and Engineering},
  year = {2026},
  volume = {14},
  number = {9},
  pages = {766},
  doi = {10.3390/jmse14090766}
}
Kostenlos starten →