User-Defined Body Gestures for Navigational Control of a Humanoid Robot.

This paper presents a study that allows users to define intuitive gestures to navigate a humanoid robot. For eleven navigational commands, 385 gestures, performed by 35 participants, were analyzed. The results of the study reveal user-defined gesture sets for both novice users and expert users. In addition, we present, a taxonomy of the user-defined gesture sets, agreement scores for the gesture sets, time performances of the gesture motions, and present implications to the design of the robot control, with a focus on recognition and user interfaces.