Knowledge-driven finite-state machines. Study case in monitoring industrial equipment

Abstract

Traditionally state machines are implemented by coding the desired behavior of a given system. This work proposes the use of ontological models to describe and perform computations on state machines by using SPARQL queries. This approach represents a paradigm shift relating to the customary manner in which state machines are stored and computed. The main contribution of the work is an ontological model to represent state machines and a set of generic queries that can be used in any knowledge-driven state machine to compute valuable information. The approach was tested in a study case were the state machines of industrial robots in a manufacturing line were modeled as ontological models and used for monitoring the behavior of these devices on real time.