Fixed-time command-filtered composite adaptive neural fault-tolerant control for strict-feedback nonlinear systems

Area of Science:

• Control Theory
• Nonlinear Systems
• Artificial Intelligence

Background:

• Strict-feedback nonlinear systems (SFNSs) often contain unknown functions and external disturbances.
• Traditional control methods struggle with the complexity and uncertainties inherent in SFNSs.
• Fault-tolerant control is crucial for maintaining system stability and performance under component failures or disturbances.

Purpose of the Study:

• To develop a fixed-time command-filtered composite adaptive neural fault-tolerant (FCCANF) control strategy for SFNSs.
• To address the challenges posed by unknown system functions and bounded disturbances.
• To ensure system internal variables remain bounded and the output converges to a small interval around zero in fixed time.

Main Methods:

• Utilizing radial basis function neural networks (RBFNNs) with serial-parallel estimation models (SPEMs) to approximate unknown functions.
• Implementing a novel fixed-time command filter and adaptive disturbance observers to manage complexity and compensate for external disturbances.
• Designing an adaptive control technique to create the FCCANF controller.

Main Results:

• Successfully addressed the issue of complexity explosion in control design.
• Effectively compensated for external disturbances.
• Demonstrated that system internal variables are bounded.
• Showcased that the output variable converges to a small interval around zero in a fixed time independent of initial conditions.
• Validated the control technique through numerical and practical examples.

Conclusions:

• The proposed FCCANF control technique is effective for strict-feedback nonlinear systems with unknown functions and disturbances.
• The method ensures finite-time convergence and robustness against uncertainties.
• The approach offers a viable solution for fault-tolerant control in complex nonlinear systems.