Decentralized Supervisory Control of Discrete Event Systems Based on Reinforcement Learning

Tatsushi YAMASAKI
Toshimitsu USHIO

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E88-A    No.11    pp.3045-3050
Publication Date: 2005/11/01
Online ISSN: 
DOI: 10.1093/ietfec/e88-a.11.3045
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Concurrent/Hybrid Systems: Theory and Applications)
Category: 
Keyword: 
discrete event systems,  decentralized control,  supervisory control,  reinforcement learning,  optimal control,  

Full Text: PDF(230KB)>>
Buy this Article


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.

open access publishing via