Neural Network-Aided Sampled-Data Control for a Class of Nonlinear Systems with Unknown Dynamics

This study addresses the stabilization control problem for a class of nonlinear systems with unknown nonlinear dynamics and unmeasurable states. To tackle this problem, we develop an artificial time-delay adaptive sampled-data control (ATASDC) framework using the artificial input delay technique. Specifically, a neural network observer is meticulously designed for state estimation, where the radial basis function neural networks (RBFNNs) are employed to approximate unknown dynamics using sampled output signals. A new Lyapunov-Krasovskii (L-K) functional is constructed to account for the sampling effects, leading to a sufficient stability condition derived via linear matrix inequalities (LMIs). The proposed method achieves semi-global practical exponential convergence of system states to a small compact set, and its effectiveness is validated through a numerical example.