Presentation 2009-01-29
Reinforcement Learning of Optimal Supervisor based on the Worst-Case Behavior
Kouji KAJIWARA, Tatsushi YAMASAKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Ramadge and Wonham proposed the supervisory control, which is a framework for logical control of discrete event systems. However, in the ordinary supervisory control, the costs for occurence and disabling of events have not been considered. This paper proposes a synthesis method of the supervisor based on the worst-case behavior of discrete event systems. We introduce the new value functions for the assigned control patterns. The new value functions are not based on the expected total rewards, but based on the most undesirable event ocurrence in the assigned control pattern. In the proposed method, the supervisor learns how to assign the control pattern based on reinforcement learning so as to maximize the value functions. We show the efficiency of the proposed method by computer simulation.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Discrete event systems / supervisory control / reinforcement learning
Paper # CST2008-49
Date of Issue

Conference Information
Committee CST
Conference Date 2009/1/22(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Concurrent System Technology (CST)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Reinforcement Learning of Optimal Supervisor based on the Worst-Case Behavior
Sub Title (in English)
Keyword(1) Discrete event systems
Keyword(2) supervisory control
Keyword(3) reinforcement learning
1st Author's Name Kouji KAJIWARA
1st Author's Affiliation Faculty of Engineering, Setsunan University()
2nd Author's Name Tatsushi YAMASAKI
2nd Author's Affiliation Faculty of Engineering, Setsunan University
Date 2009-01-29
Paper # CST2008-49
Volume (vol) vol.108
Number (no) 415
Page pp.pp.-
#Pages 6
Date of Issue