Data-Driven Repetitive Control via Noisy Data Learning: Experimental Validation on Robotic Manipulator
IEEE Transactions on Automation Science and Engineering · 2026 · 23 · 10798-10811
DOI: 10.1109/tase.2026.3699671 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Automation Science and Engineering, 2026. Verfasst von Yijia Li und 2 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Zhejiang Provincial Natural Science Foundation of China.
Abstract (Originalsprache)
This paper addresses the challenge of periodic disturbances in discrete-time nonlinear systems by proposing a direct data-driven repetitive control strategy. This framework treats the periodic disturbance term as an augmented state of the nonlinear system. A Fourier-series-based semi-definite programming (SDP) approach is developed to approximate and compensate for the periodic disturbance through disturbance rejection techniques. The proposed scheme enables learning from contaminated data while providing an estimation of the robust positive invariant (RPI) set. Numerical simulations and robotic manipulator experiments validate the effectiveness of the method in mitigating periodic interference, demonstrating its potential for practical applications in control systems engineering. Note to Practitioners—The motivation of this study stems from the necessity to develop a direct data-driven repetitive control approach for industrial robotic systems. Repetitive control (RC) offers potential to enhance system robustness against periodic disturbances. However, due to the increasing functional and structural complexity of modern industrial systems, traditional model-based repetitive control is difficult to apply directly due to high modeling difficulty. To reduce system modeling costs while improving robustness against periodic disturbances, a data-driven repetitive control strategy specifically tailored for nonlinear discrete-time systems has been developed. Through RPI analysis, an algorithm for Fourier basis function order selection is systematically designed. The effectiveness of the proposed methodology has been rigorously validated through comprehensive simulation studies and experimental implementations on robotic manipulators. This research contributes a practical and effective data-driven solution to the field of repetitive control in industrial robotics, addressing critical challenges in complex system applications.
Förderung
- National Natural Science Foundation of China
- Zhejiang Provincial Natural Science Foundation of China
Zitieren
APA
Li, Y., He, X., & Ou, X. (2026). Data-Driven Repetitive Control via Noisy Data Learning: Experimental Validation on Robotic Manipulator. IEEE Transactions on Automation Science and Engineering, 23, 10798-10811. https://doi.org/10.1109/tase.2026.3699671
Vancouver
Li Y, He X, Ou X. Data-Driven Repetitive Control via Noisy Data Learning: Experimental Validation on Robotic Manipulator. IEEE Transactions on Automation Science and Engineering. 2026;23:10798-10811. doi:10.1109/tase.2026.3699671
BibTeX
@article{Li2026,
author = {Li, Yijia and He, Xiongxiong and Ou, Xianhua},
title = {Data-Driven Repetitive Control via Noisy Data Learning: Experimental Validation on Robotic Manipulator},
journal = {IEEE Transactions on Automation Science and Engineering},
year = {2026},
volume = {23},
pages = {10798--10811},
doi = {10.1109/tase.2026.3699671}
}
EUVIMED