Ontology-based Semantic Context Modeling for Object Recognition of Intelligent Mobile Robots

Object recognitions are challenging tasks, especially invisible object recognition in changing and unpredictable robot environments. We propose a novel approach employing context and ontology to improve object recognition capability of mobile robots in realworld situations. By semantic contexts we mean characteristic information abstracted from robot sensors. We propose a method to construct semantic contexts using inferences for mobile robots to recognize objects in a more efficient way. In addition, ontology has been used for better recognizing objects using knowledge represented in the ontology. OWL (Web Ontology Language) has been used for representing object ontologies and contexts. We propose a fourlayered context ontology schema to represent perception, model, context, and activity for intelligent robots. And, axiomatic rules have been used for generating semantic contexts using OWL ontologies. Experiments are successfully performed for recognizing invisible objects based on our ontologybased semantic context model without contradictions in