Knowledge-based real time control: the use of abstraction to satisfy deadlines

From Semantic Portal Wiki

{{#vardefine:category|Publication}}{{#vardefine:templatename|i.publication}}{{#vardefine:package|smwbp_instance_templates}}

Edit

[[abstract::In this paper we address the use of model-based knowledge representation andreasoning techniques under a blackboard framework in real time computer-basedcontrol. Representing knowledge using models is one of the most powerful waysto codify the controller's knowledge about the system being controlled. Byusing a model and the current values for the monitored parameters, thecontroller can base all their process automation tasks on the outcomes of aprecise simulation of the system behavior. But models have a seriousdrawback: they have difficulty accommodating normal quantities of componentsfound in current systems. To overcome this problem when working under realtime constraints, we propose a techniques which allows a control system toachieve real time performance by means of the following computational steps:(1) starting with the basic user-defined domain model, the engineerprecomputes a hierarchy of abstract models, at increasing levels ofabstraction, using the aggregation of causal links to guide the abstractionprocess; (2) continuing off-line, the engineer estimates the worst-caseprocessing time for every model in the hierarchy; (3) on-line, the controlleruses the time constraints of the situation and the worst-case processing timesof the models to choose one model with the level of abstraction that willguarantee deadline satisfaction; and (4) the controller performs systemsimulation using the selected model. We have used the BB1 blackboardarchitecture [Hayes-Roth, 1985; 1988; 1990] to implement our work. Theresults, however, can easily be transferred to another computationalframework.|]]

Reference: {{#vardefine:pagename|knowledge-based real time control: the use of abstraction to satisfy deadlines }}

1. [[]]

bibtex

{{#vardefine:pagename|Knowledge-based real time control: the use of abstraction to satisfy deadlines }}{{#vardefine:key| }}

abstract: In this paper we address the use of model-based knowledge representation andreasoning techniques under a blackboard framework in real time computer-basedcontrol. Representing knowledge using models is one of the most powerful waysto codify the controller's knowledge about the system being controlled. Byusing a model and the current values for the monitored parameters, thecontroller can base all their process automation tasks on the outcomes of aprecise simulation of the system behavior. But models have a seriousdrawback: they have difficulty accommodating normal quantities of componentsfound in current systems. To overcome this problem when working under realtime constraints, we propose a techniques which allows a control system toachieve real time performance by means of the following computational steps:(1) starting with the basic user-defined domain model, the engineerprecomputes a hierarchy of abstract models, at increasing levels ofabstraction, using the aggregation of causal links to guide the abstractionprocess; (2) continuing off-line, the engineer estimates the worst-caseprocessing time for every model in the hierarchy; (3) on-line, the controlleruses the time constraints of the situation and the worst-case processing timesof the models to choose one model with the level of abstraction that willguarantee deadline satisfaction; and (4) the controller performs systemsimulation using the selected model. We have used the BB1 blackboardarchitecture [Hayes-Roth, 1985; 1988; 1990] to implement our work. Theresults, however, can easily be transferred to another computationalframework.

download:

• paper:
• slides: