Finite-time adaptive quantized control of stochastic nonstrict-feedback constrained nonlinear systems with multiple unmodeled dynamics

In this article, a finite-time adaptive quantized dynamic surface control problem is solved for stochastic partially nonaffine nonstrict-feedback constrained nonlinear systems with multiple unmodeled dynamics. Firstly, a variable energy function is built to handle the stochastic state unmodeled dynamics under nonstrict-feedback structure, and a normalized signal is employed to process the input unmodeled dynamics. Furthermore, a specific transformation technique is introduced to keep all states in a predefined asymmetric dynamic constrained region. Compared with the existing results, it not only avoids the circular argument, and also relaxes the condition of the constraint function. Then, the nonlinear function is approximated with the linearly parameterized neural networks. Subsequently, in order to make the tracking error reach a steady state in the finite time, with the help of the properties of the Gaussian function and dynamic surface control technique, a novel finite-time adaptive quantized controller is designed to ensure that all error signals are semi-globally practical finite-time stable and all states obey stochastic probabilistic constraints. Numerical simulation examples are provided to verify the theoretical results obtained.