Direct data-driven bipartite cooperative output consensus for heterogeneous multi-agent systems with external...

Area of Science:

• Control Theory
• Systems Engineering
• Robotics

Background:

• Investigates the bipartite cooperative output consensus (BCOC) problem in heterogeneous multi-agent systems.
• Addresses challenges posed by external disturbances and the need for model-free control strategies.

Purpose of the Study:

• To develop a direct data-driven control approach for BCOC in multi-agent systems.
• To eliminate the requirement for accurate system models by utilizing finite-time collected data.
• To derive specific regulation equations for BCOC and establish data informativity conditions.

Main Methods:

• Employs direct data-driven control, designing controllers using sampled data without a precise system model.
• Estimates the leader's system matrix using an auxiliary system's sampling data.
• Constructs a distributed observer to manage scenarios where followers lack direct leader interaction.

Main Results:

• Establishes criteria for data informativity for error system stabilization and data-driven regulation equation solutions.
• Derives specific expressions for BCOC regulation equations.
• Demonstrates the effectiveness of the proposed data-driven control strategy through a numerical simulation.

Conclusions:

• The proposed direct data-driven control method effectively solves the BCOC problem for heterogeneous multi-agent systems with external disturbances.
• The approach bypasses the need for system identification, offering a practical alternative for complex systems.
• Validated theoretical results confirm the robustness and applicability of the data-driven consensus strategy.