A Knowledge Management System for Assistive Robotics Luigi Asprino 1,2 , Aldo Gangemi 1,3 Andrea Giovanni Nuzzolese 1 , Valentina Presutti 1 , and Alessandro Russo 1 1 STLab, ISTC-CNR, Rome, Italy {name.lastname}@istc.cnr.it 2 DISI - Universit` a di Bologna, Bologna, Italy 3 LIPN, Universit´ e Paris 13, Sorbone Cit´ e, UMR CNRS, Paris, France Abstract. In this paper we demonstrate how to use an ontology net- work in order to fill the gap between knowledge and robots’ abilities. The demonstration is focused on two components of the knowledge manage- ment system of MARIO robots. These components are the MARIO On- tology Network (i.e. MON) that organises knowledge in MARIO and an Object-RDF mapper, called Lizard, that dynamically generates APIs on top of the MON to enable the interaction between software components that implement robot’s abilities and the MON itself. 1 Introduction The MARIO robot 4 is an assistive robot that has to support a set of knowledge- intensive tasks aimed at (i) helping patients affected by dementia to feel more autonomous and less lonely, (ii) supporting carers in their activity to assess the patient’s cognitive condition. Examples of knowledge-intensive tasks are the Comprehensive Geriatric Assessment (CGA) and the triggering of appropriate entertainment activities. In order to enable this tasks MARIO features a set of abilities implemented by pluggable software components. MARIO abilities, when executed, contribute to and benefit from a common knowledge base, which is modelled according to on an ontology network called MARIO Ontology Network (MON). In this paper we present two components of the Knowledge Management System of MARIO robots, i.e. (i) the MARIO Ontology Network (MON) and (ii) Lizard. The main objective of MON is to provide MARIO robots with the needed infrastructure to organise knowledge. Instead, the main objective of Lizard is to provide MARIO robots with APIs dynamically built on top of MON in order to allow the software components implementing the abilities to access, query and store the knowledge organised by the MON. The rest of the paper is organised as follows: (i) Section 2 presents the related work; (ii) Section 3 provides details about the design methodology adopted for modelling the MON (cf. Section 3.1) and the knowledge areas covered by the ontology network (cf. Section 3.2); (iii) Section 4 describes Lizard; finally, (iv) Section 5 provides the conclusions. 4 http://www.mario-project.eu/portal/