Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles

Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles

In this paper, a reinforcement learning (RL) approach is developed to solve the robust control for uncertain continuous-time linear systems. The objective is to find a feedback control law for the uncertain linear system using an online policy iteration algorithm. The robust control problem is solve...

Full description

Bibliographic Details
Main Authors: Dengguo Xu, Qinglin Wang, Yuan Li
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
Reinforcement learning
uncertain linear system
robust control
algebraic Riccati equation
Online Access:https://ieeexplore.ieee.org/document/8625521/
id doaj-48b7e34a2839476ba642ef69f669f196
record_format Article
spelling doaj-48b7e34a2839476ba642ef69f669f1962021-03-29T22:25:43ZengIEEEIEEE Access2169-35362019-01-017164311644310.1109/ACCESS.2019.28945948625521Robust Control of Uncertain Linear Systems Based on Reinforcement Learning PrinciplesDengguo Xu0https://orcid.org/0000-0003-1501-6593Qinglin Wang1Yuan Li2School of Automation, Beijing Institute of Technology, Beijing, ChinaSchool of Automation, Beijing Institute of Technology, Beijing, ChinaSchool of Automation, Beijing Institute of Technology, Beijing, ChinaIn this paper, a reinforcement learning (RL) approach is developed to solve the robust control for uncertain continuous-time linear systems. The objective is to find a feedback control law for the uncertain linear system using an online policy iteration algorithm. The robust control problem is solved by constructing an extended algebraic Riccati equation with properly defined weighting matrices for a general uncertain linear system. An online policy iteration algorithm is developed to solve the robust control problem based on RL principles without knowing the nominal system matrix. The convergence of the algorithm to the robust control solution for uncertain linear systems is proved. The simulation examples are given to demonstrate the effectiveness of the proposed algorithm. The results extend the design method of robust control to uncertain linear systems.https://ieeexplore.ieee.org/document/8625521/Reinforcement learninguncertain linear systemrobust controlalgebraic Riccati equation
collection DOAJ
language English
format Article
sources DOAJ
author Dengguo Xu
Qinglin Wang
Yuan Li
spellingShingle Dengguo Xu
Qinglin Wang
Yuan Li
Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles
IEEE Access
Reinforcement learning
uncertain linear system
robust control
algebraic Riccati equation
author_facet Dengguo Xu
Qinglin Wang
Yuan Li
author_sort Dengguo Xu
title Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles
title_short Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles
title_full Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles
title_fullStr Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles
title_full_unstemmed Robust Control of Uncertain Linear Systems Based on Reinforcement Learning Principles
title_sort robust control of uncertain linear systems based on reinforcement learning principles
publisher IEEE
series IEEE Access
issn 2169-3536
publishDate 2019-01-01
description In this paper, a reinforcement learning (RL) approach is developed to solve the robust control for uncertain continuous-time linear systems. The objective is to find a feedback control law for the uncertain linear system using an online policy iteration algorithm. The robust control problem is solved by constructing an extended algebraic Riccati equation with properly defined weighting matrices for a general uncertain linear system. An online policy iteration algorithm is developed to solve the robust control problem based on RL principles without knowing the nominal system matrix. The convergence of the algorithm to the robust control solution for uncertain linear systems is proved. The simulation examples are given to demonstrate the effectiveness of the proposed algorithm. The results extend the design method of robust control to uncertain linear systems.
topic Reinforcement learning
uncertain linear system
robust control
algebraic Riccati equation
url https://ieeexplore.ieee.org/document/8625521/
work_keys_str_mv AT dengguoxu robustcontrolofuncertainlinearsystemsbasedonreinforcementlearningprinciples
AT qinglinwang robustcontrolofuncertainlinearsystemsbasedonreinforcementlearningprinciples
AT yuanli robustcontrolofuncertainlinearsystemsbasedonreinforcementlearningprinciples
_version_ 1724191643136950272

Similar Items