A self-evolving functional-linked wavelet neural network for control applications.

Highlights: [•] The proposed self-evolving functional-linked wavelet neural network (SFWNN) can vary its structure dynamically to keep the prescribed approximation accuracy with a simple computation. [•] This paper presents an adaptive self-evolving functional-linked wavelet neural control (ASFWNC) for a class of uncertain nonlinear systems. [•] The proposed ASFWNC system is applied to a chaotic dynamic and a DC motor. [•] Simulation and experimental results verify that the proposed ASFWNC system can achieve high-precision tracking performance. [ABSTRACT FROM AUTHOR]