Nonsingular Generalized Adjustable Predefined-Time Sliding Mode Controllers With Adaptive Predefined-Time Observers for Nonlinear Dynamical Systems

作者：Xu Yang, Bin Yan, Yifeng Han, Long Chen, Hai Wang · 发表于：IEEE Transactions on Automation Science and Engineering · 年份：2026 · DOI：10.1109/tase.2026.3672987 · 被引用次数：1 · 研究领域：Adaptive Control of Nonlinear Systems、Stability and Controllability of Differential Equations、Adaptive Dynamic Programming Control

Achieving a rapid dynamic response is of critical importance in nonlinear system control. This paper develops a generalized adjustable predefined-time (GAPT) control and observation scheme for second-order nonlinear dynamical systems, aiming to overcome the limitation of non-adjustable actual convergence time in traditional predefined-time methods. Firstly, a novel GAPT system framework is proposed. This framework promotes the concept of traditional predefined time systems and introduces an adjustment parameter to flexibly adjust the actual convergence time of the system. Secondly, an adaptive generalized adjustable predefined-time observer (AGAPTO) is designed, which combines adaptive laws to ensure that the observation error converges to the specified boundary within a predefined time, while the disturbance estimation error achieves exponential convergence. Thirdly, a nonsingular generalized adjustable predefined-time sliding mode controller (NGAPTSMC) is proposed. This controller not only addresses the singularity issue of traditional predefined-time sliding modes, but also utilizes the GAPT characteristics to achieve adjustability of the actual convergence time for the arrival and switching stages. The effectiveness and superiority of the proposed method are validated through numerical simulations and real-time experiments.