Finite-Time Consensus Tracking Neural Network FTC of Multi-Agent Systems

Area of Science:

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

Background:

• Nonlinear multi-agent systems (MASs) present complex control challenges.
• Existing methods struggle with simultaneous uncertainties, actuator faults, and nonstrict feedback structures.
• Finite-time control offers faster convergence but faces design singularities.

Purpose of the Study:

• To address the finite-time consensus fault-tolerant control (FTC) tracking problem for nonlinear MASs.
• To overcome challenges posed by unknown symmetric output dead zones, actuator bias/gain faults, and unknown control coefficients.
• To develop a robust controller that guarantees finite-time convergence and bounded system signals.

Main Methods:

• Utilized neural networks (NNs) to handle unstructured uncertainties.
• Employed Nussbaum functions to manage output dead zones and unknown control directions.
• Introduced a small positive number to resolve singularity issues in finite-time backstepping design.
• Applied backstepping design and Lyapunov stability theory for controller synthesis.

Main Results:

• Developed a finite-time adaptive NN FTC controller.
• Guaranteed that the tracking error converges to a small neighborhood of zero in finite time.
• Ensured all signals within the closed-loop system remain bounded.
• Demonstrated the method's effectiveness through a physical example.

Conclusions:

• The proposed finite-time adaptive NN FTC approach effectively addresses complex fault and uncertainty scenarios in nonlinear MASs.
• The controller ensures rapid and stable system performance, with guaranteed boundedness of all states.
• This work provides a robust solution for practical applications requiring high-performance control under adverse conditions.