Command-filter-based neural networks predefined time control for switched nonlinear systems with event-triggering mechanism

The article proposes a dynamic event-triggered adaptive predefined time output feedback control technique for uncertain switching multi-input multi-output (MIMO) nonlinear systems with strict feedback forms. In contrast to previous event-triggered output feedback control, the control technique proposed in this study not only enables the system to reach steady state within a predefined time, but also further saves communication resources. Subsequently, the unpredictable states of the system are modeled using a neural network (NN) state observer. In the framework of backstepping control, an output feedback control strategy based on command filtering is proposed. Finally, the stability for a switched nonlinear system has been demonstrated using predefined time stability theory and average dwell time (ADT). The results concern this semi-global practically predefined time stabilization (SGPPTS) of all signals in the closed-loop system. Simulations and comparisons are utilized to verify the predefined time control characteristics.