A Method for Forming Mutual Beliefs for Communication through Human-robot Multi-modal Interaction

Abstract

This paper describes a method of multimodal language processing that reflects experiences shared by people and robots. Through incremental online optimization in the process of interaction, the user and the robot form mutual beliefs represented by a stochastic model. Based on these mutual beliefs, the robot can interpret even fragmental and ambiguous utterances, and can act and generate utterances appropriate for a given situation.