Adaptive neural network asymptotic tracking control for nonstrict feedback stochastic nonlinear systems

The adaptive neural network asymptotic tracking control issue of nonstrict feedback stochastic nonlinear systems is studied in our article by adopting backstepping algorithm. Compared with the existing research, the hypothesis about unknown virtual control coefficients (UVCC) is overcome in the control design. By using the bound estimation scheme and some smooth functions, associating with approximation-based neural network, the asymptotic tracking controller is recursively constructed. With the aid of Lyapunov function and beneficial inequalities, the asymptotic convergence character and stability with stochastic disturbance and unknown UVCC can be ensured. Finally, the theoretical finding is verified via a simulation example. (C) 2021 Published by Elsevier Ltd.

Automated summary

In this article, the issue of adaptive neural network asymptotic tracking control for nonstrict feedback stochastic nonlinear systems is studied using the backstepping algorithm. Compared with previous research, the difficulty of unknown virtual control coefficients in control design is overcome. The recursive construction of the asymptotic tracking controller is achieved through the use of bound estimation scheme, smooth functions, and approximation-based neural network, ensuring asymptotic convergence character and stability with stochastic disturbance and unknown UVCC with the help of Lyapunov function and beneficial inequalities. This theoretical finding is verified through a simulation example.