Ontologenius: A long-term semantic memory for robotic agents

Abstract

In this paper we present Ontologenius, a semantic knowledge storage and reasoning framework for autonomous robots. More than a classic ontology software to query a knowledge base and a first-order internal logic as it can be done for web-semantics, we propose with Ontologenius features adapted to a robotic use including human-robot interaction. We introduce the ability to modify the knowledge base during execution, whether through dialogue or geometric reasoning, and keep these changes even after the robot is powered off. Since Ontologenius was developed to be used by a robot which interacts with humans, we have endowed the system with ability to perform attributes and properties generalization and with the possibility to model and estimate the semantic memory of a human partner and to implement theory of mind processes. This paper presents the architecture and the main features of Ontologenius as well as examples of its use in robotics applications.

The Director Task: a Psychology-Inspired Task to Assess Cognitive and Interactive Robot Architectures

“…The semantic knowledge base is in the form of an ontology as it allows a rich expressiveness, a standardisation of the representation among several architectures, and reasoning capabilities. We chose the software Ontologenius [20] to manage it. A reason for this choice is that it is fully adapted to HRI applications by representing the robot's knowledge and the estimation of the partners' knowledge separately, which refers to the psychological concept of the "self-other distinction" as coined in joint action study [21].…”

Extending Referring Expression Generation through shared knowledge about past Human-Robot collaborative activity

“…The original algorithm and its extended version have been run over all the 77 entities inheriting from the "Object" class, representing physical entities. The knowledge base is managed using the Ontologenius system [28]. The extension has a negligible impact when no task is described in the ontology, with an average execution time of 1.04ms for the original algorithm versus 1.16ms for the extension.…”

Efficient, Situated and Ontology based Referring Expression Generation for Human-Robot collaboration

“…3) Ontology in the context of Human-Robot Interaction: Since we are dealing with HRI applications, it is pertinent to maintain a knowledge base per agent in order to implement theory of mind decisional mechanisms. This is provided by the system we use, Ontologenius [15] which is an efficient framework designed for robotic applications and which allows to manage several ontologies corresponding to the knowledge of the robot and to its estimated knowledge of its human partners.…”

“…The knowledge base is managed using the Ontologenius 4 system [15]. It uses a custom internal structure to store and manipulate assertions as triplets, and offers reasoning capabilities in the form of plugins.…”