Authors
Yifei Zhu and Alexander Torres and Zane Aloia and Tom Williams
Venue
IEEE/RSJ International Conference on Intelligent Robots and Systems
Publication Year
2025
Robots that use gestures in conjunction with speech can achieve more effective and natural communication with human teammates, however, not all robots have capable and dexterous arms. Augmented Reality technology has effectively enabled deictic gestures for morphologically limited robots in prior work, however, the design space of AR-facilitated iconic gestures remains under-explored. Moreover,
existing work largely focuses on closed-world context, where all referents are known a priori. In this work, we present a human-subject study situated in an open-world context, and compare the task performance and subjective perception associated with three different iconic gesture designs (anthropomorphic, non-anthropomorphic, deictic-iconic) against previously studied abstract gesture design. Our quantitative and qualitative results demonstrate that deictic iconic gestures (in which a robot hand is shown pointing to a visualization of a target referent) outperforms all other gestures on all metrics – but that non-anthropomorphic iconic gestures (where a visualization of a
target referent appears on its own) is overall most preferred by users. These results represent a significant step forward to enabling effective human-robot interactions in realistic large-scale open-world environments.
