Designing Interfaces for Synthesizing Robot Dialogue and Behaviors

Successful human-robot interaction relies heavily on the robot’s programming, making it essential for its design to follow human behavioral patterns as closely as possible. Despite the complexity of human communication and dialogue, we have identified key patterns of interaction, such as that can be applied to a variety of scenarios, including conversation, collaboration, instruction, interview, and storytelling.

By building an interface that encompasses these fundamental patterns, our goal is to make it easier for designers to create human-like dialogues. We also aim to make a user-friendly authoring environment that even non-experts can use, with drag-and-drop sequencing and a visually simple layout. Currently our authoring environment consists of three sections, the control panel, the pattern library, and interaction timeline, which will we can continue to develop and modify through our research and evaluations. With more simplified programming, we can empower anyone—designer or not—to craft a successful human-robot dialogue.

Please contact Allison Sauppé at asauppe{at}cs.wisc.edu for any questions regarding this project.

This research is supported by National Science Foundation #1149970

Publications

Sauppé, A., & Mutlu, B. Design Patterns for Exploring and Prototyping Human-Robot Interactions. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI 2014). May 2014. Toronto, Canada