Adaptive Neural Dynamic Surface Output Feedback Control for Nonlinear Full States Constrained Systems

In order to solve the control problem of uncertain nonlinear systems with state constraints, a dynamic surface output feedback control technology based on Radial Basis Function (RBF) neural networks state observer is proposed. The state observer is designed to estimate the unknown state of the systems by using the approximation characteristics of RBF neural networks, and to constrain the system state by using the Barrier Lyapunov Function (BLF). Based on the backstepping control, a first-order low-pass filter is introduced to design a dynamic surface control (DCS), which solves the “differential explosion” phenomenon that can easily occur in backstepping control. Finally, the stability of the closed-loop system which is confirmed by the Lyapunov method guarantees the semi-globally uniformly ultimately boundedness (SGUUB) of all the signals. The effectiveness of the methods that the boundedness of the tracking errors, the observer states and the controllers can be guaranteed, and good control performance could be achieved is shown by simulation results.