Indiana University and the Luddy School of Informatics, Computing, and Engineering made a big splash at the 19th Annual ACM/IEEE International Conference on Human Robot Interaction that ran March 11-15 in Boulder, Colorado. It is Human-Robot Interaction’s premiere conference.
Luddy School faculty and students, led by Selma Šabanović, professor of Informatics and associate dean for faculty affairs, presented four papers as oral presentations and four short papers as posters at the conference.
David Crandall, Luddy Professor of Computer Science and director of Luddy Artificial Intelligence Center, helped lead three of the papers.
Šabanović’s paper, “Constructing a Social Life with Robots, From Design Patterson to Interaction Ritual Chains,” co-authored with former Luddy Master’s student Waki Kamino and Cornell professor Malte F Jung, received a best-paper recognition award at the selective international conference which attracted almost 800 attendees and featured 87 full-paper presentations.
The paper focused on showing how people’s interactions with each other and with robots in various everyday activities lead them to experience robots as social others and for robots to become an accepted part of daily life.
Robots play ever-increasing roles in everyday life as well as in science and industry. In Japan in particular, people socialize with robots and use them for functional tasks. They take robots home; they wake up with them; they return home from work and interact with them; and they take them out on play dates with other robots and their owners.
“People are creating all these new daily rituals of living in which robots play social roles,” Šabanović says.
As part of their research, Kamino went to Japan last summer and collected observations and interviews with company representatives who sell commercial robots and commercial robot owners who used them in their daily lives.
Šabanović said their study suggests the general criticism that interacting with robots keeps you from spending time with people is wrong.
“Interactions with robots are part of interactions with other people. Instead of the robot taking you away from human society, it’s an activity that people do together. They are socializing around the robot, not just with it.”
Added Kamino: “By showing how robot owners have added robots to their existing social routines shared with other friends and family members, we’d like to stress that people, not design features, make robots social.”
Crandall and Šabanović co-authored another paper, “Give it Time: Longitudinal Panels Scaffold Older Adults’ Learning and Robot Do-Design”, along with Luddy Ph.D. student Long-Jing Hsu, postdoc Weslie Khoo, research engineer Kyrie Jig Amon, researcher Manasi Swaminathan, M.S. student Hiroki Sato, and retired IU anthropology professor Phillip B. Stafford.
The paper addressed the team’s research into designing a social robot for older adults, especially those living with dementia. The team used an innovative co-design process in which the people who would later use the robot -- older adults and their caregivers -- actively participated in its design.
Every two weeks, a panel of older adults came to IU to evaluate the robot’s latest interactive features. In the process, the older adults learned about robotics and AI in order to be able to provide more meaningful feedback.
“The focus was not so much about the robot,” Crandall said, “as it was on the people and how they learned to talk about AI and robotics.”
Another of their papers, “If (Your Name) Can Code, So Can You! End-User Robot Programming for Non-Experts,” presented a new technique to help people program social robots’ behavior.
Finding new ways for people to engage in robot design is important, Šabanović said.
“Many people want to try something out. How do you do that quickly How do you determine if a new feature is working or not? That feedback is important in design. The team has worked out a way for the robot to be programable in this quick, easy, testable way.”
Two of Šabanović’s papers dealt with privacy issues that can arise with robots in the home. One investigated privacy perceptions of adults, while another investigated perceptions and concerns of teenagers, a population rarely featured in HRI research. The latter paper, co-authored with Ph.D. student Leigh Levinson and Informatics Professor Christena Nippert-Eng, shows that teens understand privacy in the context of their social relationships, so it is important to them to understand how the robot may mediate between them and their friends, family and teachers.
“Robots often collect visual data with cameras,” Šabanović says. “Robots can also pick up your speech, how you move, or if you’re if touching the robot. All this information goes somewhere. Sometimes, it stays in the robot. Sometimes, it goes to a company or to a friend or family member of the user.
“We need to think about the issues of collecting and sharing data. How do these connect with the law, with policies, with people’s needs and understanding of what robots can do? There are a lot of open questions.”
To help address these questions, Levinson and Sabanović co-organized a workshop at the conference focused on privacy issues in social robotics, centered on information collected by robots during their interactions with humans.
Regarding this work, Levinson said, “The workshop was a wonderful way to invigorate the study of privacy that is so critical to successful human-robot interaction. These papers and a continued effort to focus on privacy in the community are a staple to the research we are doing at our IU Lab.”
Šabanović has seen the Luddy presence at the conference grow over the 15 years since she’s been at IU. Luddy has emerged to join Carnegie Mellon, Yale University, the University of Wisconsin-Madison and other institutions as a “force to be reckoned with” as far as the number of participants, papers presented and workshops directed.
“Now we have a whole contingent of IU faculty, students, and alumni in the HRI field,” she says. “It’s noticeable. That’s exciting.”
In addition to giving IU students the opportunity to present their work on an international stage, Šabanović said the conference provided inspiration for the students to talk with leading experts in the field.
“Working in your office or lab is one thing,” she said. “Seeing other people doing similar work from all over the world and interacting with people who care about what you have to say and who are interested in what you’re doing is very motivational. It can give you a year’s worth of energy to further pursue your research.”
The Luddy School emphasizes humanity as well as technology. The work presented at the conference, Crandall said, “was the embodiment of that idea.”
“One of the focuses of the Luddy School is on human-centered computing and human-centered AI in particular,” he said. “Research into Human-Robot Interaction is an essential part of that vision, and I’m so proud of Selma and our team that we have so many contributions from Luddy there.”