Output Containment Control for Heterogeneous Linear Multiagent Systems With Fixed and Switching Topologies.

Area of Science:

• Control Theory
• Networked Systems
• Robotics

Background:

• Multiagent systems require coordinated control for complex tasks.
• Output containment control ensures follower agents remain within a desired region defined by leaders.

Purpose of the Study:

• To investigate and solve the output containment control problem for heterogeneous linear multiagent systems.
• To achieve exponential convergence of follower outputs into the convex hull of leader outputs.

Main Methods:

• Derivation of a necessary condition based on the internal model principle.
• Design of a dynamic controller for exponential containment.
• Construction of an optimal control law using an algebraic Riccati equation.
• Application of a reinforcement learning algorithm for online optimal control without system dynamics knowledge.

Main Results:

• A necessary condition for output containment is established.
• Exponential output containment is achieved for general network topologies (fixed and dynamic).
• A stabilizing optimal control law is derived.
• The reinforcement learning approach successfully solves the optimal control problem online.

Conclusions:

• The proposed methods effectively solve the output containment control problem for heterogeneous linear multiagent systems.
• The study validates theoretical findings through simulations, demonstrating practical applicability.