Heterogeneous Context-Aware Robots Providing a Personalized Building Tour

Abstract

Existing robot guides offer a tour of a building, such as a museum or science centre, to one or more visitors. Usually the tours are predefined and lack support for dynamic interactions between the different robots. This paper focuses on the distributed collaboration of multiple heterogeneous robots (receptionist, companion) guiding visitors through a building. Semantic techniques support the formal definition of tour topics, the available content on a specific topic, and the robot and person profiles including interests and acquired knowledge. The robot guides select topics depending on their participants' interests and prior knowledge. Whenever one guide moves into the proximity of another, the guides automatically exchange participants, optimizing the amount of interesting topics. Robot collaboration is realized through the development of a software module that allows a robot to transparently include behaviours performed by other robots into its own set of behaviours. The multi-robot visitor guide application is integrated into an extended distributed heterogeneous robot team, using a receptionist robot that was not originally designed to cooperate with the guides. Evaluation of the implemented algorithms presents a 90% content coverage of relevant topics for the participants.