Command filter-based adaptive neural finite-time control for stochastic nonlinear systems with time-varying full-state constraints and asymmetric input saturation

This article studies a tracking control problem for a class of stochastic nonlinear systems with time-varying full-state constraints and asymmetric input saturation. Firstly, the Gauss Error Function is introduced to solve the difficulty arising from the saturation nonlinearity. Meanwhile, to overcome the problem of calculating explosion caused by the repeated differentiation of the virtual control signals, a finite-time command filter with a compensation mechanism is developed. Combining the neural networks' approximation ability and the backstepping technique, an adaptive neural finite-time control strategy is proposed for the considered system by constructing the time-varying barrier Lyapunov function. Under the proposed control strategy, it is guaranteed that all signals are bounded in a sense of mean square, the output of the system can track the reference signal within a finite time and all states will not violate the constraints. Furthermore, the stability of the closed-loop system is analysed based on the stochastic finite-time stability theory. Finally, a simulation example verifies the effectiveness of the proposed control strategy.

Automated summary

This article proposes an adaptive neural finite-time control strategy for stochastic nonlinear systems, which combines neural network approximation and backstepping technique, constructs a time-varying barrier Lyapunov function, and solves the difficulty arising from saturation nonlinearity. With the proposed control strategy, it is guaranteed that system signals are bounded, the reference signal is tracked within a finite time, and system states do not violate constraints.