Event-Triggered Reference Governors for Collisions-Free Leader-Following Coordination under Unreliable Communication Topologies

• Published in: IEEE transactions on automatic control Vol. 69; no. 4; pp. 1 - 14
• Main Authors: An, Liwei, Yang, Guang-Hong, Deng, Chao, Wen, Changyun
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.04.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Closed loops; Collision avoidance; Collisions; Coordination; Feedback control; Governors; Leader-following coordination; Liapunov functions; Lyapunov methods; multi-agent networks; Multiagent systems; Nonlinear systems; Safety; Switches; systems security; Target tracking; Topology; Tracking errors; Transient analysis
• ISSN: 0018-9286; 1558-2523
• DOI: 10.1109/TAC.2023.3291654

Reference governors are add-on control schemes which enforce state constraints on pre-stabilized systems by modifying, whenever necessary, the reference. This paper studies the design schemes of reference governors for collisions-free leader-following coordination (CF-LFC) of nonlinear multi-agent systems under unreliable communications. The "unreliability" is characterized as irregular topology switchings (e.g., due to communication failures, adversarial attacks, or other agents' blockages), which will often cause large transient coordination errors and further result in interagent collisions. To address it, an event-triggered reference governor is proposed which dominates agents to perform the leader-following task if the tracking error meets the closed-loop system performance requirement, and otherwise the agents will temporarily "forget" the task and focus on avoiding collisions. By constructing heterogeneous Lyapunov functions, it is shown that the reference governor can achieve the CF-LFC despite the presence of topology switchings. Also, for a special class of nonlinear systems with strict-feedback form, a non-event-triggered reference governor is further proposed by reconstructing the controller with a high-order integrator of barrier function as the control gain to adaptively adjust the transient tracking errors.