Enriching Robots with Ontological-Cognitive Knowledge (E-ROCK)
half-day tutorial at the
26th IEEE International Symposium on Robot and Human Interactive Communication (IEEE RO-MAN 2017)
Lisbon, Portugal, August 28, 2017
Thanks to the many participants for making this tutorial enjoyable and interactive. We had just scratched the surface. There is plenty of space for a follow-up E-Rock in the future.
Laboratory for Applied Ontology, ISTC CNR
Via alla cascata 56C, 38123 Trento (Italy)
stefano.borgo [ at ] cnr.it
The tutorial aims to introduce the audience to the use of ontology for modelling the common-sense knowledge needed by a robotic agent to understand the environment and to interact with it. The goal is to study how to organise the robot’s information system with knowledge of the world captured from the human perspective. This approach makes possible to develop human-robot interactions on the basis of a shared common-sense/cognitive understanding of the actual situation. We introduce and discuss the notions and relationships that are needed to “understand” a generic scenario and also show how to structure an ontology to organize such knowledge. In particular, we will focus on how to understand and model engineering functions, actions, contexts and environments.
What is an ontology and why do we need it in robotics?
How to read a simple scenario from a human’s perspective and how to furnish this view to a robot.
How to integrate common-sense/cognitive knowledge about objects, goals, rules, functions and actions.
What are contexts and environments? How to model them? For what purposes?
How to integrate knowledge, goal, action, environment and context.
The tutorial is situated in the research area across ontology and robotics and is at the introductory level. No specific knowledge is required. Familiarity with formal languages can be helpful in some example. Students, researchers and practitioners are welcome.