🤖This page has been translated by AI and may contain inaccuracies. Please refer to the original content for clarification.

On the stability condition of active disturbance rejection control with time-varying bandwidth observer
Depeng Song
Sen Chen
Wenchao Xue
Zhiliang Zhao
Abstract:With the growing adoption of artificial intelligence algorithms and neural networks, online learning and adaptive methods for updating the bandwidth have become increasingly prevalent. However, the conditions required to ensure closed-loop stability when employing a time-varying bandwidth, as well as the supporting mathematical foundations, remain insufficiently studied. This paper investigates the stability condition for active disturbance rejection control (ADRC) with a time-varying bandwidth extended state observer (ESO). A new stability condition is derived, which means that the upper bound of rate of change for ESO bandwidth should be restricted. Moreover, under the proposed condition, the closed-loop stability of ADRC with a time-varying bandwidth observer is rigorously proved for nonlinear uncertainties. In simulations, the necessity of the proposed condition is illustrated, demonstrating that the rate of change of ESO bandwidth is crucial for closed-loop stability.
Keywords:Active disturbance rejection controlTime-varying bandwidthExtended state observerClosed-loop stabilityRate of change
Publication Date:2025-08-30
Online Publishing Date:2025-12-09(First online date of this platform, not the publication date of the document)
Pages:15( 464-478 )
Control Theory and Technology

Control Theory and Technology

EICSCD
ISSN:2095-6983
Year, Vol.(Issue):2025,23(3)