Robust coalitional model predictive control with plug-and-play capabilities

This article presents a distributed implementation of a model predictive controller with information exchange to manage a distributed networked system of coupled dynamic subsystems. We propose a coalitional control method, where local controllers coalesce into clusters to improve performance, as a t...

ver descrição completa

Detalhes bibliográficos
Autores: Masero Rubio, Eva, Baldivieso-Monasterios, Pablo R., Maestre Torreblanca, José María, Trodden, Paul A.
Tipo de documento: artigo
Estado:Versão publicada
Data de publicação:2023
País:España
Recursos:Universidad de Sevilla (US)
Repositório:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/145359
Acesso em linha:https://hdl.handle.net/11441/145359
https://doi.org/10.1016/j.automatica.2023.111053
Access Level:Acceso aberto
Palavra-chave:Distributed networked systems
Model predictive control
Control by clustering
Plug-and-play events
Tubes
Descrição
Resumo:This article presents a distributed implementation of a model predictive controller with information exchange to manage a distributed networked system of coupled dynamic subsystems. We propose a coalitional control method, where local controllers coalesce into clusters to improve performance, as a tool to solve plug-and-play problems. Our main contribution is a tube-based coalitional approach that employs online optimized invariant sets. These sets are instrumental in guaranteeing recursive feasibility and stability when faced with plug-and-play operations, i.e., subsystems joining or leaving the network. We also explore the inherent robustness properties to absorb disturbances not covered by the tubes without the need to group local controllers. Finally, the simulation results show the benefits of our proposed control method.