A maximum-entropy-inspired parser
NAACL 2000 Proceedings of the 1st North American chapter of the Association for Computational Linguistics conference
Hi-index | 0.00 |
To allow a robot to function in unstructured environments, a control architecture should be flexible and should facilitate the creation and representation of arbitrarily complex sequences of actions. Such an architecture should also facilitate simple and natural interaction between a robot and its human collaborators. Human language has both representational flexibility and naturalness to human operators; in this paper we show how the structures of natural language can serve as the basis of an effective robot control architecture.