Predefined-time distributed optimization and anti-disturbance control for nonlinear multi-agent system with neural...

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Robotics

Background:

• Distributed optimization is crucial for multi-agent systems.
• Handling unknown nonlinearities and disturbances is a significant challenge.
• Predefined-time control offers faster convergence than traditional methods.

Purpose of the Study:

• To develop a predefined-time distributed optimization strategy for nonlinear multi-agent systems.
• To address challenges posed by unknown nonlinear functions and external disturbances.
• To ensure optimal consensus trajectories and accurate system tracking within a fixed time.

Main Methods:

• A two-layer hierarchical control framework is proposed.
• A predefined-time distributed estimator generates optimal consensus trajectories.
• Neural networks approximate unknown nonlinearities, and a disturbance observer estimates external disturbances.
• A neural-network-based anti-disturbance sliding mode control ensures trajectory tracking.

Main Results:

• The proposed hierarchical control framework guarantees predefined-time convergence and stability, verified by Lyapunov analysis.
• The system trajectories successfully track optimal trajectories within the predefined time.
• Simulations on robotic arm and mobile robot models demonstrate the method's effectiveness.

Conclusions:

• The presented hierarchical control approach effectively achieves predefined-time distributed optimization for nonlinear multi-agent systems.
• The integration of neural networks and disturbance observers enhances robustness against uncertainties.
• The method provides a reliable solution for complex multi-agent coordination tasks in robotics.