Data-Driven Dynamic Triggered Formation Control for Multi-Agent Systems

This paper presents a data-driven dynamic triggered formation control approach for multi-agent systems. The development of a dynamic event-triggered mechanism that adaptively adjusts triggering thresholds based on the relationship between tracking errors and triggering errors, reducing network communication load. A data-driven approach is used in the design of state feedback controllers for scenarios where system model information is inaccessible due to privacy protection requirements, and a comprehensive stability analysis using Lyapunov theory to prove the closed-loop stability of the system. Simulation results on a multi-UAV formation demonstrate that the proposed method significantly reduces communication frequency compared to time-triggered mechanisms while maintaining tracking performance, providing an effective solution for distributed control of MASs without precise modeling requirements.