Pelc, Mariusz (2011) Reconfigurable autonomic control systems (Rekonfigurowalne autonomiczne systemy sterowania). PAK, 57 (5). pp. 460-462. ISSN 0032-4110

PDF (Journal article) 11_70.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

In this paper a very relevant issue related to reconfigurable autonomic real-time control systems is undertaken. Designers of this kind of systems have to balance on the one side the ability of their operation with virtually no (or with minimum) human intervention, but on the other side, they have to implement a mechanism for easy and efficient reconfiguration in response to changing environmental conditions (context awareness). Although there are many techniques that may potentially be used to implement autonomic behaviour (such as Artifical Neural Networks, Fuzzy Logic, etc) actually only policy-based computing seems to give the system designers enough freedom when it comes to specification of how the system should behave in response to the environmental changes. This ease results from the fact, that each policy is written using a Policy Description Language (PDL) which can be optimized for the given problem area (various control systems applications, business applications, etc) As each PDL (for example AGILE PDL) offers besides a specific structure also a set of keywords strictly related to the problem domain, the system designers can easily express relations between context variables changes (these reflect the environment / context changes) and the requested changes to the system behaviour in a descriptive way. This is a very big advantage of policy-based computing over the alternative technologies. Policies themselves constitute the software component "logic" and because they (and thus thelogic itself) can easily be replaced with newer (more optimized) versions, then in the result the same software component may behave completely different under the same environmental conditions. Policy-based computing is the ideal candidate technology to support reconfiguration of autonomic systems as this technology is not resource hungry, so it may be sucessfully applied to the embedded systems domain.

Item Type:	Article
Uncontrolled Keywords:	computer control systems, policy-based computing
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Pre-2014 Departments:	School of Computing & Mathematical Sciences School of Computing & Mathematical Sciences > Department of Smart Systems Technologies
Related URLs:	Publisher Organisation
Last Modified:	14 Oct 2016 09:18
URI:	http://gala.gre.ac.uk/id/eprint/7096

Actions (login required)

View Item

Downloads