Decentralized adaptive neural asymptotic control of switched nonlinear interconnected systems with predefined tracking performance

摘要

This article aims at the problem of decentralized adaptive neural asymptotic tracking for switched non-linear interconnected systems with unknown strong interconnections and predefined transient perfor-mance. Technically, unknown strong interconnection terms are handled by intrinsic properties of the basis function vector. In addition, unlike prescribed performance bound control, a new error -dependent transformation with a time-varying function is proposed, which completely circumvents the initial condition-dependence problem. With such transformation and a class of integral bounded -based tuning functions, a decentralized adaptive neural asymptotic control strategy is established so that closed-loop stability can be preserved, and the output tracking error not only asymptotically converges to zero but also evolves within the prescribed boundary. Finally, illustrative examples validate the obtained results. CO 2022 Published by Elsevier B.V.

关键词