Robust Model Reference Adaptive Control Based on Reproducing Kernel Hilbert Spaces.

Saved in:

Bibliographic Details
Title:	Robust Model Reference Adaptive Control Based on Reproducing Kernel Hilbert Spaces.
Authors:	Wang, Haoran¹ (AUTHOR), Kurdila, Andrew J.¹ (AUTHOR), L'Afflitto, Andrea² (AUTHOR) a.lafflitto@vt.edu, Oesterheld, Derek³ (AUTHOR), Stilwell, Daniel J.³ (AUTHOR)
Source:	International Journal of Adaptive Control & Signal Processing. Jun2025, Vol. 39 Issue 6, p1128-1148. 21p.
Subjects:	Distributed parameter systems, Adaptive control systems, Ordinary differential equations, Hilbert space, Robust control
Abstract:	This article introduces native space embedding for robust adaptive control of ordinary differential equations that contain vector‐valued functional uncertainties in a reproducing kernel Hilbert space (RKHS). The proposed approach is based on a two‐phase method for analyzing and designing adaptive controllers. In the first phase, a limiting distributed parameter system (DPS), which describes the ideal closed‐loop system's performance, is introduced. The limiting DPS is not realizable in practice since it evolves in a generally infinite‐dimensional space. In the second phase, consistent finite‐dimensional approximations of the DPS are introduced to determine realizable controllers. Uniform ultimate bounds on the trajectory tracking error dynamics are derived for the functional uncertainty classes contained in the native space. These bounds are derived in terms of the power function of the RKHS or in terms of the fill distance of centers that define the scattered basis for approximations. Two numerical examples demonstrate the applicability of the proposed results. [ABSTRACT FROM AUTHOR]
	Copyright of International Journal of Adaptive Control & Signal Processing is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	This article introduces native space embedding for robust adaptive control of ordinary differential equations that contain vector‐valued functional uncertainties in a reproducing kernel Hilbert space (RKHS). The proposed approach is based on a two‐phase method for analyzing and designing adaptive controllers. In the first phase, a limiting distributed parameter system (DPS), which describes the ideal closed‐loop system's performance, is introduced. The limiting DPS is not realizable in practice since it evolves in a generally infinite‐dimensional space. In the second phase, consistent finite‐dimensional approximations of the DPS are introduced to determine realizable controllers. Uniform ultimate bounds on the trajectory tracking error dynamics are derived for the functional uncertainty classes contained in the native space. These bounds are derived in terms of the power function of the RKHS or in terms of the fill distance of centers that define the scattered basis for approximations. Two numerical examples demonstrate the applicability of the proposed results. [ABSTRACT FROM AUTHOR]
ISSN:	08906327
DOI:	10.1002/acs.3997