Hooman Hedayati, a computer science PhD student based in the ATLAS Institute’s Interactive Robotics and Novel Technologies (IRON) Lab, participated in a prestigious summer internship at Microsoft Research in Redmond, Washington, where he worked on teaching robots social skills in group conversations. His research focused on helping robots detect F-formations, group conversations that happen when “two or more people sustain a spatial and orientational relationship in which the space between them is one to which they have equal, direct and exclusive access.”
“For us, detecting F-formations is easy, and we don’t think about it,” says Hedayati. “You know how many are in your conversational group, and you know how to position yourself in respect to others. But this task is not easy for robots.”
During the internship, Hooman and his Microsoft Research mentor, Sean Andrist, worked on developing an algorithm to help robots detect those in the same conversational group as the robot. The two plan to publish a paper about their findings.
“It was a great feeling to be surrounded by top scientists and legends in my field,” Hedayati says.