Output-Feedback Adaptive Neural Controller for Uncertain Pure-Feedback Nonlinear Systems Using a High-Order Sliding...

Area of Science:

• Control Systems Engineering
• Nonlinear Dynamics
• Artificial Intelligence

Background:

• Adaptive neural controllers for pure-feedback nonlinear systems often rely on complex dynamic surface control (DSC) or backstepping schemes.
• These traditional methods result in lengthy control laws and complicated stability analyses.
• Research on output-feedback neural controllers for these systems remains limited.

Purpose of the Study:

• To propose a novel adaptive neural output-feedback controller for SISO nonaffine pure-feedback nonlinear systems.
• To simplify the control law and stability analysis compared to existing methods.
• To address the limited research in output-feedback neural control for this system class.

Main Methods:

• Reformulating the nonlinear system into the Brunovsky form to simplify control design.
• Integrating a high-order sliding mode observer.
• Utilizing a single radial-basis function neural network with universal approximation capabilities.

Main Results:

• The proposed controller successfully avoids complex adaptive backstepping or DSC schemes.
• The reformulation into Brunovsky form significantly simplifies the control law.
• The combination of the observer and neural network guarantees closed-loop system stability.

Conclusions:

• A novel and simplified adaptive neural output-feedback controller is presented for SISO nonaffine pure-feedback nonlinear systems.
• The controller ensures closed-loop system stability in the Lyapunov sense.
• This approach offers a more tractable solution compared to traditional DSC or backstepping methods.