Distributed Approximate Aggregative Optimization of Unknown Pure-Feedback Systems With Sampled Neighbor Information.

Area of Science:

• Control Systems Engineering
• Networked Systems
• Nonlinear Dynamics

Background:

• Distributed aggregative optimization (DAO) is crucial for networked systems.
• Extending DAO to high-order nonlinear systems with unknown dynamics presents significant challenges.

Purpose of the Study:

• To develop a novel framework for distributed aggregative optimization in high-order nonlinear systems with unknown pure-feedback dynamics.
• To address control challenges in directed and unbalanced networks.

Main Methods:

• Introduction of auxiliary aggregative variables for integrating agent and neighbor information.
• Development of a smoothing function for progressive variable updates.
• Application of dynamic average consensus principles and a pivotal theorem to transform the DAO problem into a regulation problem.
• Design of a control law using prescribed performance functions for approximate optimization under bounded disturbances.

Main Results:

• Successfully extended DAO methodologies to high-order nonlinear systems.
• Demonstrated the transformation of the DAO problem into a solvable regulation problem.
• Validated the proposed control scheme's effectiveness through numerical examples for agents with unknown dynamics and disturbances.

Conclusions:

• The proposed approach effectively solves the distributed aggregative optimization problem for high-order nonlinear systems.
• The method offers a robust solution for complex networked control systems with unknown dynamics and disturbances.