Distributed Supervision Strategies for Cyber-Physical Systems With Varying Network Topology

• 成果类型: Article
• 署名作者: Tedesco, Francesco; Casavola, Alessandro
• 署名单位: University of Calabria
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2024.3362884
• 发表日期: 2024
• 页码: 5408-5423
• 关键词: Couplings; Cyber-physical systems; TOPOLOGY; decentralized control; Computer architecture; Vehicle dynamics; stability analysis; Cyber physical systems; command governor; distributed reference management; plug and play control; varying network topology

摘要

This article presents a novel distributed constrained monitoring strategy for cyber-physical systems that focuses on enforcing point-in-time set-membership coordination constraints among the subsystem evolutions. The proposed scheme extends previous solutions by allowing the handling of time-varying dynamic interconnections, including the online addition/removal of subsystems (plug-and-play) during normal system operations. This approach ensures global coordination of the subsystems while effectively handling corresponding changes in the underlying constraint topology. The coordination is achieved by determining a distributed feasible set-point for each subsystem when the nominal set-point becomes unfeasible due to topological changes in the system, resulting in constraints violation. In order to address this challenge, this article generalizes the distributed command governor (CG) theory to handle online requests for structural changes in the system and/or in the coordination constraints. An add-on module is provided for each local CG to process the coordination constraints and appropriately schedule the topology changes when certain formal conditions are satisfied. This ensures that the coordination constraints are satisfied even during the topological changes. Case studies are presented by evaluating the proposed strategy on the coordination of a formation of moving vehicles and a water distribution network. These simulations are used to evaluate the effectiveness and performance of the approach.