An industrial application using blackboard architecture
Kerem Burak Tünay · Akademik Açık Erişim (Işık Üniversitesi) · 2006
This thesis implements control architecture for goal-driven blackboard systems.The architecture is based on searching a general goal tree by diminishing into sub-goal trees.The aim is to develop a problem solving architecture in the AI space via blackboard system.The basic elements of the architecture are goals, policies, strategies, facts, methods, and knowledge sources.The basic control loop employs a bidding mechanism to determine the knowledge source to be executed at the current cycle.A policy is a local scheduling criterion which guides to bidding process and it indicates which of the attributes of the knowledge sources are relevant in this process.A strategy is a global scheduling criteria such as depthfirst, breadth-first etc.A method is a partially complete general goal tree structure representing high level knowledge on how to solve a problem.The architecture employs a control blackboard, and separate knowledge sources for the control problem and for representing the domain knowledge.A production planning application is developed using this architecture.Both C++ and ABAP languages were used to implement this application.