RBF: An object and constraint oriented reasoning blackboard framework for knowledge based applications
Date of Award
Doctor of Philosophy (PhD)
Electrical Engineering and Computer Science
Chilukuri K. Mohan
Knowledge-based applications, Automated reasoning, Reasoning, Blackboard
Computer Sciences | Systems Architecture
This dissertation presents a new reasoning framework called the Object and Constraint oriented Reasoning Blackboard Framework for Knowledge Based Applications (RBF), with a blackboard architecture enhanced by a unique goal directed, object oriented, and constraint based problem solving capability. The goal directed reasoning used in RBF utilizes a formal approach specialized to deal with knowledge sources, panels, objects and constraints. A new approach is provided to facilitate the addition and dynamic invocation of application specific constraint solvers and thereby flexibly enhance the inference engine. RBF provides a blackboard control mechanism which can be used in several ways, for example to compute shortest paths for problem solving through intermediate paths or to apply specialized goal directed reasoning or any inference procedure to decide the next step in the problem solving process. RBF also provides a uniform mechanism for applying rule sets using both forward and backward chaining. The utility of RBF is shown with reference to two applications: implementation of invertible translators for transforming image formats, and interpretation of sonar data. A specialized approach is introduced that is they basis for declarative invertible mapping of object models, with the use of constraints involving object trees. The specialized goal-directed inference approach works with blackboard elements and constraints for control driven problem solving such as testing hypotheses or looking for a discrepancy as in the sonar data interpretation application.
Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.
Chidella, Jagannadha, "RBF: An object and constraint oriented reasoning blackboard framework for knowledge based applications" (2002). Electrical Engineering and Computer Science - Dissertations. Paper 118.