Neural Network-Based Adaptive Fixed-Time Terminal Sliding Mode Control for Unknown Nonlinear Systems

The adaptive fixed-time terminal sliding mode control (TSMC) problem will be discussed for unknown nonlinear strict-feedback systems. Neural networks are used to estimate the unknown nonlinearities existing in the systems. When the control scheme is designed, the dynamic surface control (DSC) method is applied to avert the derivation of the virtual control signal and a novel hyperbolic tangent filter is developed. Subsequently, the adaptive control method is introduced to proposed a terminal sliding surface (TSS). Consequently, an adaptive TSMC controller is developed. Using the Lyapunov stability theory, we prove that the tracking error inclines to the TSS and a small interval near zero, and the closed-loop signals are bounded within fixed time. Finally, the availability of the proposed scheme is demonstrated by the simulation example.