Neural Preassigned Performance Control for State-Constrained Nonlinear Systems Subject to Disturbances.

Area of Science:

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

Background:

• State-constrained nonlinear systems (NSs) are challenging to control due to inherent limitations.
• Exogenous disturbances degrade system performance and stability.
• Achieving predefined performance within a finite time requires advanced control strategies.

Purpose of the Study:

• To develop a finite-time neural predefined performance control (PPC) for state-constrained nonlinear systems (NSs).
• To ensure tracking errors meet predefined performance indicators (PPIs) within a finite time.
• To enhance disturbance rejection and system robustness.

Main Methods:

• Integration of predefined-time performance functions (PTPFs) and barrier Lyapunov functions (BLFs) for time-varying constraints.
• Application of nonlinear disturbance observer techniques (NDOT) for disturbance estimation and compensation.
• Utilizing dynamic surface control (DSC) for recursive controller design.

Main Results:

• A novel finite-time neural adaptive PPC strategy was successfully devised.
• The proposed controller ensures full-state constraints are satisfied.
• The closed-loop system demonstrates semi-globally practically finite-time stability (SPFS) and achieves desired PPIs.
• Simulation results validate the approach's effectiveness and robustness against disturbances.

Conclusions:

• The developed finite-time neural adaptive PPC strategy effectively handles state constraints and disturbances in nonlinear systems.
• The method guarantees predefined performance within a finite time, improving control accuracy and stability.
• The approach offers a viable solution for complex control problems requiring high robustness and performance.