Ontology Learning of New Concepts Combining Textural Knowledge, Visual Analysis, and User Interaction

Abstract

The continuous development of new technology has been contributing to making service robots an upcoming reality. Robots need to adapt to changing environments when interacting with people is fundamental. Hence, service robots at home need learning capabilities to acquire new knowledge and merge it with their own. In this system, a robot is provided with an essential feature to adapt inside a home environment. We focus on the learning of new ontological concepts oriented to service robot applications. We propose combining textural knowledge, visual analysis and user interaction to determine the correct placement of the new concept in the ontology structure. We aim to make the robot able to extend its ontological knowledge as needed. We conducted a set of experiments to show the applicability of the presented method and the advantage of having the objects conceptualized in ontological knowledge. The experiments consisted of two parts: concept learning experiments and experiments with an integrated robot system. In the concept learning experiments, the robot had to conceptualize a set of new objects in its ontological knowledge. While in the integrated robot system experiments, the robot was asked to search and find the new objects learned.INDEX TERMS Concept learning, ontology learning, robot learning, human-robot interaction.