Composite Learning Fuzzy Control of Stochastic Nonlinear Strict-Feedback Systems

This article investigates the composite learning fuzzy control for a class of stochastic nonlinear strict-feedback systems subject to dynamics uncertainty. The fuzzy logic system is built to model the unknown system nonlinearity. The highlight is that different from previous studies using only tracking error for fuzzy weight updating, the accuracy of fuzzy learning is emphasized in this study. The serial-parallel estimation model with fuzzy approximation and gain compensation is constructed to acquire the prediction error such that the composite fuzzy updating law is designed with more accurate feedback information. The stochastic stability analysis ensures the uniformly ultimate boundedness of the system signals in mean square. Through the simulation tests on a numerical example with different stochastic disturbances and one-link manipulator dynamics, it is proved that the proposed composite learning scheme can solve the system uncertainty effectively and make the closed-loop system track the reference command with satisfactory accuracy.

Automated summary

This study investigates a composite learning fuzzy control for a class of stochastic nonlinear strict-feedback systems, emphasizing the accuracy of fuzzy learning. By constructing a serial-parallel estimation model, a more accurate feedback information composite fuzzy updating law is designed. Through simulation tests, it is proved that the proposed scheme can effectively solve system uncertainty.