Constrained quadratic control of discrete-time hidden Markovian jump linear systems: the state feedback and static output scenarios.

• Ano de defesa: 2021
• Autor(a) principal: Andres Zabala, Yeison
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Controle estocástico; Controle ótimo; Desigualdades; Hidden Markov Models; Jump linear systems; Linear matrix inequalities; Matriciais lineares; Optimal control; Sistemas lineares com salto; Stochastic control
• Link de acesso: https://www.teses.usp.br/teses/disponiveis/3/3139/tde-22022022-104313/

In this thesis 2 global scenarios are considered for the constrained optimal control for hidden MJLS, where it is assumed that the controller only has access to a detector which emits signals b(k) providing information on the Markov parameter (k). The State Feedback Quadratic Control (first scenario) and Static Output Feedback Control (second scenario) problems are studied. Hence, they are obtained, via LMIs approach, feedback linear optimal controls so that the respective closed loop systems are stochastically stabilized, an upper-bound for the quadratic cost is minimized, and the constraints on the norm of the state and control variables are satis ed. The Finite Horizon and the Infinite Horizon cases as well as the maximization of the estimate of the domain of an invariant set for a fixed upper-bound of the cost function are also addressed. Finally, some numerical examples are presented for the purpose of illustrating the obtained results.