Colonel Roche

This paper presents how extraction, representation and use of symbolic knowledge from real-world perception and human-robot verbal and non-verbal interaction can actually enable a grounded and shared model of the world that is suitable for later high-level tasks like dialogue understanding, sym-bolic task planning or reactive supervision. We show how the anchoring process itself fundamentally relies on both the situated and embodied nature of human-robot interactions. We present an implementation, including a specialized sym-bolic knowledge representation system based on Description Logics, and experiments on several robotic platforms that demonstrate these cognitive capabilities. Keywords human-robot interaction, symbolic knowledge anchoring, on