Supervisor Localization: A Top-Down Approach to Distributed Control of Discrete-Event Systems

Journal Article · Thu Mar 05 00:00:00 EST 2009 · AIP Conference Proceedings

DOI:https://doi.org/10.1063/1.3106490· OSTI ID:21293329

Cai, K; Wonham, W M ^[1]

---

1. Systems Control Group, Department of Electrical and Computer Engineering, University of Toronto, 10 King's College Road, Toronto, ON, M5S 3G4 (Canada)

A purely distributed control paradigm is proposed for discrete-event systems (DES). In contrast to control by one or more external supervisors, distributed control aims to design built-in strategies for individual agents. First a distributed optimal nonblocking control problem is formulated. To solve it, a top-down localization procedure is developed which systematically decomposes an external supervisor into local controllers while preserving optimality and nonblockingness. An efficient localization algorithm is provided to carry out the computation, and an automated guided vehicles (AGV) example presented for illustration. Finally, the 'easiest' and 'hardest' boundary cases of localization are discussed.

• OSTI ID: 21293329
• Journal Information: AIP Conference Proceedings, Vol. 1107, Issue 1; Conference: CISA'09: 2. Mediterranean conference on intelligent systems and automation, Zarzis (Tunisia), 23-25 Mar 2009; Other Information: DOI: 10.1063/1.3106490; (c) 2009 American Institute of Physics; Country of input: International Atomic Energy Agency (IAEA); ISSN 0094-243X
• Country of Publication: United States
• Language: English

Related Subjects

71 CLASSICAL AND QUANTUM MECHANICS
GENERAL PHYSICS
ALGORITHMS
ARTIFICIAL INTELLIGENCE
CALCULATION METHODS
CONTROL
DESIGN
NEURAL NETWORKS
ROBOTS
VEHICLES