Decentralized Optimal Control for Linear Stochastic Systems with Control Signals subject to Unknown Noises

Decentralized strategies have been extensively applied to LQ optimal control problems, whereas, stochastic systems with unknown random parameters have not been comprehensively studied. In this paper, we consider a class of stochastic systems with a decentralized configuration consisting of multiple...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE International Conference on Control and Automation (Print) s. 186 - 191
Hlavní autoři: Zhang, Zhaorong, Xu, Juanjuan, Fu, Minyue, Li, Xun
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 18.06.2024
Témata:
Approximation algorithms
Automation
Gaussian noise
Optimal control
Q-learning
Riccati equations
Stochastic systems
ISSN:1948-3457
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Decentralized strategies have been extensively applied to LQ optimal control problems, whereas, stochastic systems with unknown random parameters have not been comprehensively studied. In this paper, we consider a class of stochastic systems with a decentralized configuration consisting of multiple controllers which have access to Gaussian noises with unknown statistical information. The stabilizing and optimal control strategies are acquired by designing a novel stochastic approximation algorithm recursively evaluating the zero points of certain matrix equations, which is confirmed to be equivalent with solving the corresponding Riccati equations. The proof of convergence and boundness of the proposed algorithm is presented.
ISSN:1948-3457
DOI:10.1109/ICCA62789.2024.10591873