The paper proposes a knowledge-based framework for mobile autonomous robots. It exploits data annotation for semantic-based context description. High-level event/situation detection and action decision are performed through a semantic matchmaking approach, supporting approximate matches and relevance-based ranking. The framework was fully implemented in a prototype built with off-the-shelf components, validated in a Search And Rescue (SAR) case study and evaluated in an early performance analysis, supporting the feasibility of the proposal. The work demonstrates novel analysis methods on data extracted by inexpensive sensors can yield useful results without requiring hefty computational resources.
Knowledge-based sensing/acting in mobile autonomous robots / Ruta, Michele; Scioscia, Floriano; Loseto, Giuseppe; DI SCIASCIO, Eugenio. - (2017), pp. 422-427. (Intervento presentato al convegno The First IEEE International Conference on Robotic Computing (IRC 2017) tenutosi a Taichung, Taiwan nel April 10-12, 2017) [10.1109/IRC.2017.83].
Knowledge-based sensing/acting in mobile autonomous robots
RUTA, Michele;SCIOSCIA, Floriano;LOSETO, Giuseppe;DI SCIASCIO, Eugenio
2017-01-01
Abstract
The paper proposes a knowledge-based framework for mobile autonomous robots. It exploits data annotation for semantic-based context description. High-level event/situation detection and action decision are performed through a semantic matchmaking approach, supporting approximate matches and relevance-based ranking. The framework was fully implemented in a prototype built with off-the-shelf components, validated in a Search And Rescue (SAR) case study and evaluated in an early performance analysis, supporting the feasibility of the proposal. The work demonstrates novel analysis methods on data extracted by inexpensive sensors can yield useful results without requiring hefty computational resources.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
