• DocumentCode
    2193847
  • Title

    Strong co-observability conditions for decentralized supervisory control of discrete event systems

  • Author

    Takai, Shigemasa ; Ushio, Toshimitsu

  • Author_Institution
    Dept. of Opto-Mechatronics, Wakayama Univ., Japan
  • Volume
    5
  • fYear
    2001
  • fDate
    2001
  • Firstpage
    4098
  • Abstract
    We study nonblocking decentralized supervisory control of discrete event systems. A notion of normality which is a weaker version of strong decomposability has been defined in our previous work (2001). However, a nonblocking decentralized supervisor which achieves the supremal Lm (G)-closed, controllable, and normal sublanguage may be still unnecessarily restrictive. The main purpose of the paper is to synthesize a more permissive nonblocking decentralized supervisor. We introduce stronger versions of C&P co-observability and D&A co-observability, called strong C&P co-observability and strong D&A co-observability, respectively. Strong C&P co-observability and strong D&A co-observability are weaker than normality. Moreover, strong D&A co-observability is preserved under union. So a nonblocking decentralized supervisor exists under the AND rule for the supremal Lm(G)-closed, controllable, and strongly D&A co-observable sublanguage. We present a procedure for computing this supremal sublanguage
  • Keywords
    automata theory; control system synthesis; decentralised control; discrete event systems; formal languages; observability; automata theory; control system synthesis; decentralized supervisory control; discrete event systems; formal languages; normality; strong C&P co-observability; strong D&A co-observability; supremal Lm(G)-closed controllable normal sublanguage; Control system synthesis; Control systems; Discrete event systems; Humans; Observability; Optimal control; Supervisory control;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Decision and Control, 2001. Proceedings of the 40th IEEE Conference on
  • Conference_Location
    Orlando, FL
  • Print_ISBN
    0-7803-7061-9
  • Type

    conf

  • DOI
    10.1109/.2001.980821
  • Filename
    980821