Neural-Network-Based Adaptive Tracking Control for Nonlinear Multiagent Systems: The Observer Case

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Nonlinear Dynamics

Background:

• Multiagent systems often face challenges with immeasurable states and external disturbances.
• Adaptive tracking control is crucial for coordinating system behavior.

Purpose of the Study:

• To develop neural-network (NN)-based adaptive tracking control for high-order nonlinear multiagent systems.
• To address immeasurable state variables and unknown external disturbances.
• To ensure follower systems accurately track the leader's output.

Main Methods:

• Utilizing radial basis function neural networks (RBF NNs) for adaptive control.
• Implementing composite disturbance and state observers.
• Applying the backstepping technique for stability analysis.
• Extending the control to an event-triggered framework to conserve resources.

Main Results:

• Achieved semiglobal uniform ultimate boundedness for all closed-loop system signals.
• Demonstrated that follower outputs ultimately track the leader's output.
• Validated the proposed adaptive event-triggered tracking control protocol through a numerical example.

Conclusions:

• The proposed NN-based adaptive control effectively manages nonlinear multiagent systems under uncertainty.
• The event-triggered approach offers a resource-efficient solution for adaptive tracking control.
• The method provides robust tracking performance and system stability.