Visual Servo Tracking Control for Robot Manipulator via Koopman Operator-Based Disturbance Observer
IEEE Transactions on Automation Science and Engineering · 2026 · 23 · 10478-10491
DOI: 10.1109/tase.2026.3698243 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Automation Science and Engineering, 2026. Verfasst von Yongzheng Cong und 3 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Anhui Province Science and Technology Innovation Breakthrough Plan, Basic Research Program of Jiangsu.
Abstract (Originalsprache)
The primary objective of this paper is to address the visual servo tracking control problem for robot manipulator. Based on the visual servo control system structure, the system is artificially divided into two parts: the outer-loop (visual feedback control loop) and the inner-loop (robot tracking control loop). First, for the outer loop system, a visual servo pixel kinematics model is established based on the perspective projection equation, and a virtual linear velocity controller for the camera is designed. Second, based on the robot’s Jacobian matrix, the virtual linear velocity of the outer-loop is transformed into the robot joint angular velocity signal (the desired signal for the inner-loop). Then, combining the robot’s dynamic characteristics, a finite-time trajectory tracking controller is designed to ensure that the robot can track the desired joint angular velocity signal in a finite time. Additionally, considering the presence of external disturbances in the robot system, a koopman operator-based disturbance observer is designed to enhance the robustness of the inner loop control system. Finally, numerical simulations and experimental results validate the feasibility of the proposed theoretical algorithms. Note to Practitioners—Visual servo control is widely used in robotic systems operating in complex task environments, such as flexible automated production lines and medical robotic systems. However, in practical applications, robot systems are often affected by external disturbances and internal model uncertainties, which may degrade control accuracy. To address this issue, this paper develops a disturbance observer with active learning capability based on a model- and data-driven framework using Koopman operator theory. The proposed observer can estimate unknown disturbances in real time and improve control performance. Moreover, the method does not rely on an accurate system model and is easy to implement, making it suitable for integration into existing visual servo control frameworks to enhance the performance and robustness of robotic systems in complex environments.
Förderung
- National Natural Science Foundation of China
- Anhui Province Science and Technology Innovation Breakthrough Plan
- Basic Research Program of Jiangsu
- Jiangsu Provincial Scientific Research Center of Applied Mathematics
Zitieren
APA
Cong, Y., Du, H., Zhu, W., & Wen, G. (2026). Visual Servo Tracking Control for Robot Manipulator via Koopman Operator-Based Disturbance Observer. IEEE Transactions on Automation Science and Engineering, 23, 10478-10491. https://doi.org/10.1109/tase.2026.3698243
Vancouver
Cong Y, Du H, Zhu W, Wen G. Visual Servo Tracking Control for Robot Manipulator via Koopman Operator-Based Disturbance Observer. IEEE Transactions on Automation Science and Engineering. 2026;23:10478-10491. doi:10.1109/tase.2026.3698243
BibTeX
@article{Cong2026,
author = {Cong, Yongzheng and Du, Haibo and Zhu, Wenwu and Wen, Guanghui},
title = {Visual Servo Tracking Control for Robot Manipulator via Koopman Operator-Based Disturbance Observer},
journal = {IEEE Transactions on Automation Science and Engineering},
year = {2026},
volume = {23},
pages = {10478--10491},
doi = {10.1109/tase.2026.3698243}
}
EUVIMED