Further reduction of minimal first-met bad markings for the computationally efficient synthesis of a maximally permissive controller

Autor:	Daniel Yuh Chao, Gaiyun Liu
Rok vydání:	2015
Předmět:	Reduction (complexity) Supervisor Control and Systems Engineering Control theory Flexible manufacturing system State (computer science) Deadlock Deadlock prevention algorithms Integer programming Computer Science Applications Mathematics
Zdroj:	International Journal of Control. 88:1423-1428
ISSN:	1366-5820 0020-7179
DOI:	10.1080/00207179.2014.991940
Popis:	To date, research on the supervisor design for flexible manufacturing systems focuses on speeding up the computation of optimal (maximally permissive) liveness-enforcing controllers. Recent deadlock prevention policies for systems of simple sequential processes with resources (S3PR) reduce the computation burden by considering only the minimal portion of all first-met bad markings (FBMs). Maximal permissiveness is ensured by not forbidding any live state. This paper proposes a method to further reduce the size of minimal set of FBMs to efficiently solve integer linear programming problems while maintaining maximal permissiveness using a vector-covering approach. This paper improves the previous work and achieves the simplest structure with the minimal number of monitors.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::b66f4fcc53edfdc7a8ab76067c4e7062 https://doi.org/10.1080/00207179.2014.991940 Zobrazit plný text záznamu