RT Ontology development and human preference learning for assistive robotic service system

In service robotics systems, understanding the relationship between environmental objects and user intention is the key feature to provide suitable services according to context. RT Ontology has shown to be an efficient technique to represent this relationship, yet it contains non-context information. In this paper, we propose a novel method to develop the RT Ontology automatically and a learning algorithm to connect the context-free model of RT Ontology with human preference. Resulting system is capable of providing assistive contextual services to user, as well as learning human action preference.