Maintaining Awareness of the Focus of Attention of a Conversation

We explored online reinforcement learning techniques to find good policies to control the orientation of a mobile robot during social group conversations. In this scenario, we assumed that the correct behavior for the robot should convey attentiveness to the focus of attention of the conversation. Thus, the robot’s goal was to turn towards the speaker. Our results from tests in a simulated environment showed that a new state representation that we designed for this problem can be used to find good motion policies for the robot. These policies show potential to generalize across interactions with different numbers of people and various levels of sensing noise.


We thank the Walt Disney Corporation for supporting this research effort, as well as Christoph Dann and Emma Brunskill for their feedback and suggestions.

M. Vázquez, A. Steinfeld, S. E. Hudson. Maintaining Awareness of the Focus of Attention of a Conversation: A Robot-Centric Reinforcement Learning Approach. Proc. of the 2016 IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), 2016.
» Finalist of the Best Paper Award (Technology category) and the RSJ/KROS Distinguished Interdisciplinary Research Award