Action-centered communication between a human instructor and a robot requires new methods of language processing. In this paper, we present an approach which considers natural language processing not in isolation but as a distributed competence of an artificial agent. We introduce a hybrid architecture which combines a behavior-oriented base system with a deliberative component. The architecture integrates language, perception and action in order to allow the processing of action directives on different levels of complexity and to achieve given tasks in a flexible and responsive manner.
© 2001-2026 Fundación Dialnet · Todos los derechos reservados