A Multilayer Graph for Multiagent Formation and Trajectory Tracking Control Based on MPC Algorithm.

Area of Science:

• Robotics
• Control Systems Engineering
• Network Science

Background:

• Multiagent systems require sophisticated control for coordinated movement and task execution.
• Existing interaction network models often lack extensibility and adaptability.
• Trajectory tracking and formation control are critical for autonomous systems in dynamic environments.

Purpose of the Study:

• To develop a novel control strategy for multiagent systems focusing on formation and trajectory tracking.
• To introduce a flexible multilayer graph representation for multiagent interactions.
• To ensure collision and obstacle avoidance while maintaining formations and tracking reference paths.

Main Methods:

• A novel multilayer graph is proposed to model multiagent system interactions.
• A potential function-based formation control law is developed for autonomous formation, maintenance, collision, and obstacle avoidance.
• Model Predictive Control (MPC) is applied to a virtual leader (barycenter of the formation) for trajectory tracking.

Main Results:

• The proposed multilayer graph facilitates extensible interaction networks.
• The control law successfully achieves autonomous formation, maintenance, and avoidance maneuvers.
• The MPC scheme enables the virtual leader to accurately track reference trajectories while agents maintain formation.
• The integrated approach fulfills formation maintenance and trajectory tracking in constrained spaces.

Conclusions:

• The developed multilayer graph and control strategies effectively address formation and trajectory tracking challenges in multiagent systems.
• The method demonstrates robustness and effectiveness under various environmental constraints and trajectories.
• This work provides a significant advancement in the control of coordinated autonomous systems.