Decentralized Supervisory Control of Discrete Event Systems Based on Reinforcement Learning

Tatsushi YAMASAKI; Toshimitsu USHIO

doi:10.1093/ietfec/e88-a.11.3045

Decentralized Supervisory Control of Discrete Event Systems Based on Reinforcement Learning

Tatsushi YAMASAKI, Toshimitsu USHIO

Full Text Views

0

Share
Cite this

Summary :

A supervisor proposed by Ramadge and Wonham controls a discrete event system (DES) so as to satisfy logical control specifications. However a precise description of both the specifications and the DES is needed for the control. This paper proposes a synthesis method of the supervisor for decentralized DESs based on reinforcement learning. In decentralized DESs, several local supervisors exist and control the DES jointly. Costs for disabling and occurrence of events as well as control specifications are considered. By using reinforcement learning, the proposed method is applicable under imprecise specifications and uncertain environment.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E88-A No.11 pp.3045-3050

Publication Date: 2005/11/01

Publicized

Online ISSN

DOI: 10.1093/ietfec/e88-a.11.3045

Type of Manuscript: Special Section PAPER (Special Section on Concurrent/Hybrid Systems: Theory and Applications)

Category

Cite this

Copy

Tatsushi YAMASAKI, Toshimitsu USHIO, "Decentralized Supervisory Control of Discrete Event Systems Based on Reinforcement Learning" in IEICE TRANSACTIONS on Fundamentals, vol. E88-A, no. 11, pp. 3045-3050, November 2005, doi: 10.1093/ietfec/e88-a.11.3045.
Abstract: A supervisor proposed by Ramadge and Wonham controls a discrete event system (DES) so as to satisfy logical control specifications. However a precise description of both the specifications and the DES is needed for the control. This paper proposes a synthesis method of the supervisor for decentralized DESs based on reinforcement learning. In decentralized DESs, several local supervisors exist and control the DES jointly. Costs for disabling and occurrence of events as well as control specifications are considered. By using reinforcement learning, the proposed method is applicable under imprecise specifications and uncertain environment.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e88-a.11.3045/_p

Copy

@ARTICLE{e88-a_11_3045,
author={Tatsushi YAMASAKI, Toshimitsu USHIO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Decentralized Supervisory Control of Discrete Event Systems Based on Reinforcement Learning},
year={2005},
volume={E88-A},
number={11},
pages={3045-3050},
abstract={A supervisor proposed by Ramadge and Wonham controls a discrete event system (DES) so as to satisfy logical control specifications. However a precise description of both the specifications and the DES is needed for the control. This paper proposes a synthesis method of the supervisor for decentralized DESs based on reinforcement learning. In decentralized DESs, several local supervisors exist and control the DES jointly. Costs for disabling and occurrence of events as well as control specifications are considered. By using reinforcement learning, the proposed method is applicable under imprecise specifications and uncertain environment.},
keywords={},
doi={10.1093/ietfec/e88-a.11.3045},
ISSN={},
month={November},}

Copy

TY - JOUR
TI - Decentralized Supervisory Control of Discrete Event Systems Based on Reinforcement Learning
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 3045
EP - 3050
AU - Tatsushi YAMASAKI
AU - Toshimitsu USHIO
PY - 2005
DO - 10.1093/ietfec/e88-a.11.3045
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E88-A
IS - 11
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - November 2005
AB - A supervisor proposed by Ramadge and Wonham controls a discrete event system (DES) so as to satisfy logical control specifications. However a precise description of both the specifications and the DES is needed for the control. This paper proposes a synthesis method of the supervisor for decentralized DESs based on reinforcement learning. In decentralized DESs, several local supervisors exist and control the DES jointly. Costs for disabling and occurrence of events as well as control specifications are considered. By using reinforcement learning, the proposed method is applicable under imprecise specifications and uncertain environment.
ER -