K-filter observer based adaptive multiple tan-combined command-filtered control for nonlinear systems with unknown asymmetric nonlinear dead-zone input

This article investigates adaptive neural network command-filtered control for nonlinear systems, incorporating tan-based error compensation signals and addressing unknown asymmetric nonlinear dead-zone input. An innovative combination of compensated state errors and tan-type barrier Lyapunov functions (TTBLF) are introduced. Simultaneously, by incorporating the newly designed tan-based error compensation signals, full state constraints are achieved. Subsequently, the unknown asymmetric nonlinear dead-zone model is established and transformed into a linear asymmetric dead-zone model for further processing without conventional inverse dead-zone compensation. Additionally, a neural filter state observer is developed, allowing for direct measurement of all states. Finally, the RLC circuit is presented to illustrate the effectiveness and superiority of the proposed method.