EUVIMEDEUVIMED entdecken
Crossref

VigilMPC: Certified Online Tube Model Predictive Control for Safe Real-Time Neuromodulation

Yu Mao, Zhichao Liang, Junxiang Zhang, Quanying Liu, Tongxin Li

IEEE Control Systems Letters · 2026 · 10 · 1831-1836

DOI: 10.1109/lcsys.2026.3711599 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Control Systems Letters, 2026. Verfasst von Yu Mao und 4 weiteren Autor:innen. Die Arbeit nennt 10 Literaturangaben. Gefördert von: Shenzhen Natural Science Foundation in Basic Research Fund, 1+1+1 CUHK-CUHK(SZ)-GDSTC Joint Collaboration Fund, Guangdong Basic and Applied Basic Research Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Closed-loop neuromodulation demands safe adaptive stimulation: requirements are strongly patient-specific and may vary across symptoms and operating conditions, while unsafe updates can reduce efficacy or induce side effects. We propose VigilMPC, a certified online tube model predictive control (MPC) framework with online model learning. Unlike adaptive MPC safety-certification schemes that recertify feasibility by updating uncertainty descriptions within a prescribed model class, VigilMPC certifies each learned nonlinear predictor update before deployment, accounting for its effect on both multi-step rollouts and tube tightenings. A candidate predictor is accepted only if the shifted backup plan remains tube-feasible; otherwise, the previously certified predictor and stored plan are retained. Under standard tube-MPC assumptions, VigilMPC guarantees recursive feasibility and constraint satisfaction. Simulations on a switching benchmark and a Jansen–Rit neural mass model demonstrate safe adaptation with prototype-level timing for biomarker-level neuromodulation.

Förderung

  • Shenzhen Natural Science Foundation in Basic Research Fund
  • 1+1+1 CUHK-CUHK(SZ)-GDSTC Joint Collaboration Fund
  • Guangdong Basic and Applied Basic Research Foundation
  • National Natural Science Foundation of China
  • PengCheng Peacock Scientific Research Fund
  • Guangdong Key Lab of Math Foundations for AI
  • Shenzhen Key Lab of Crowd Intelligence Empowered Low-Carbon Energy Network

Zitieren

APA

Mao, Y., Liang, Z., Zhang, J., Liu, Q., & Li, T. (2026). VigilMPC: Certified Online Tube Model Predictive Control for Safe Real-Time Neuromodulation. IEEE Control Systems Letters, 10, 1831-1836. https://doi.org/10.1109/lcsys.2026.3711599

Vancouver

Mao Y, Liang Z, Zhang J, Liu Q, Li T. VigilMPC: Certified Online Tube Model Predictive Control for Safe Real-Time Neuromodulation. IEEE Control Systems Letters. 2026;10:1831-1836. doi:10.1109/lcsys.2026.3711599

BibTeX
@article{Mao2026,
  author = {Mao, Yu and Liang, Zhichao and Zhang, Junxiang and Liu, Quanying and Li, Tongxin},
  title = {VigilMPC: Certified Online Tube Model Predictive Control for Safe Real-Time Neuromodulation},
  journal = {IEEE Control Systems Letters},
  year = {2026},
  volume = {10},
  pages = {1831--1836},
  doi = {10.1109/lcsys.2026.3711599}
}
Kostenlos starten →