Dynamic-gain neural network observer based prescribed performance backstepping sliding mode control of uncertain...

Area of Science:

• Control Systems Engineering
• Nonlinear Dynamics
• Artificial Intelligence in Engineering

Background:

• Nonlinear systems often face challenges with unknown disturbances and unmeasurable states, hindering precise control.
• Traditional observers require manual tuning, which is impractical for systems with time-varying uncertainties.
• Achieving specific transient and steady-state performance bounds is crucial for reliable real-world applications.

Purpose of the Study:

• To develop a robust control framework for nonlinear systems with unknown disturbances and uncertainties.
• To enhance system reliability by ensuring tracking errors meet pre-specified performance requirements.
• To improve the adaptability of state observers for nonlinear systems.

Main Methods:

• A prescribed performance backstepping sliding mode control (SMC) framework was designed.
• A dynamic-gain neural network observer was incorporated to estimate unmeasurable states and handle uncertainties.
• An integral nonsingular fast terminal SMC (INFTSMC) strategy was integrated with prescribed performance control (PPC).
• Lyapunov theory was used for stability analysis.

Main Results:

• The dynamic-gain observer adaptively adjusts its gain in real-time, eliminating the need for precise tuning.
• The integrated INFTSMC and PPC strategy ensures tracking errors meet defined transient and steady-state requirements.
• The proposed control method effectively manages system dynamics while the observer compensates for nonlinearities.
• Simulation results validated the effectiveness and robustness of the developed control framework.

Conclusions:

• The proposed control framework offers a robust and reliable solution for nonlinear systems with unknown disturbances.
• The combination of a dynamic-gain neural network observer and INFTSMC with PPC significantly enhances control performance.
• The method demonstrates practical applicability by ensuring adherence to performance specifications and system stability.