R3D3: The Rolling Receptionist Robot with Double Dutch Dialogue

Jeroen Linssen, Mariet Theune

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    4 Citations (Scopus)
    1 Downloads (Pure)


    We discuss the design of R3D3, a rolling receptionist robot with the ability to conduct 'double Dutch dialogues': dialogues (in Dutch) that involve, besides a human user, both a robot and a virtual human. R3D3 is intended to assist people when they visit shops, museums, or other establishments by acting as a host or receptionist. In the R3D3 project, we investigate how users can interact with the robot through natural language and nonverbal behaviour. R3D3 uses computer vision to determine user characteristics and combines this with automated speech recognition to analyse users' intentions. The robot and virtual human complement each other's affordances for verbal and nonverbal behaviour. Our preliminary studies have shown that `double Dutch dialogues' require careful attention management through verbal and nonverbal behaviours by the robot and the virtual human, and that a receptionist robot should take the initiative in conversations to fulfil its intentions.
    Original languageEnglish
    Title of host publicationProceedings of the Companion of the 2017 ACM/IEEE International Conference on Human-Robot Interaction
    PublisherAssociation for Computing Machinery (ACM)
    Number of pages2
    ISBN (Electronic)978-1-4503-4885-0
    Publication statusPublished - Mar 2017
    Event12th ACM/IEEE International Conference on Human-Robot Interaction, HRI 2017: Smart Interaction - Vienna, Austria
    Duration: 6 Mar 20179 Mar 2017
    Conference number: 12


    Conference12th ACM/IEEE International Conference on Human-Robot Interaction, HRI 2017
    Abbreviated titleHRI
    Internet address


    • Robot Gaze
    • Robot
    • Dialogue


    Dive into the research topics of 'R3D3: The Rolling Receptionist Robot with Double Dutch Dialogue'. Together they form a unique fingerprint.

    Cite this