Home

Social interaction involves a large number of patterned behaviors that people choose and combine to achieve particular communicative goals. To achieve fluent and effective human-robot communication, robots must seamlessly integrate different social behaviors for a given interaction context. The Robot Behavior Toolkit is an implementation of a framework that contextualizes (1) an expandable set of social scientific findings on human social behavior in (2) an interaction model that supports human activity based on Activity Theory. The Toolkit is developed to be extensible and flexible to various sensor devices and robotic platforms. To this end, we integrated the Toolkit with the ROS infrastructure. The Toolkit is written in Python and adopts XML as the main format for information sharing and storage.