Eleanor Avrunin

Email address: eavrunin at cs dot cmu dot edu


I am a graduate student at the Robotics Institute at Carnegie Mellon University, advised by Professor Reid Simmons. I completed the MS en route to the PhD in Spring 2015, and am currently taking a leave of absence.

I graduated from Yale University in May, 2011. As a student there, I majored in Computer Science and worked in the Social Robotics Laboratory, under the direction of Professor Brian Scassellati.


I am particularly interested in social robotics, and how human perceptions of robots influence interactions with robots.

My current project is studying social navigation: how does the way a robot moves through a space and navigates around people affect their perceptions of it, and how can we make robots move in socially-appropriate ways. I am investigating the ways that the linguistic theory of Common Ground can be used to describe the nonverbal communication involved in indicating one's intended path.

E. Avrunin and R. Simmons. Socially-Appropriate Approach Paths Using Human Data. In Proceedings of the 23rd IEEE International Symposium on Robot and Human Interactive Communication. August, 2014. 1037-1042. DOI

For service robots operating in indoor environments, the crucial task of navigation is often complicated by the presence of people. Simply treating humans in the environment as additional (often moving) obstacles can violate the complex set of social rules by which people navigate around each other. In contrast, emulating human behavior and navigating in a socially appropriate manner could positively affect people's comfort with a robot's presence and motion. We present a method of generating social paths for a robot to approach a person based on a small amount of human data. We also conducted a study in which a robot approached participants using both these social paths and straight-line, nonsocial paths. We found that both approaches were rated comparably when the robot approached from the participant's front or side, but the social approach was significantly preferred when the robot came from behind the participant.

E. Avrunin and R. Simmons. Using Human Approach Paths to Improve Social Navigation. In Proceedings of the 8th International Conference on Human-Robot Interaction. March, 2013. 73-74. DOI

Here we have presented a small study collecting human social navigation data, and a method for adding an approximation of human social rules for approaching others to a robot's autonomous navigation by using a model of those data. For a robot to adhere to those social rules when it moves through human environments would make its navigation more predictable and familiar to humans, who have years of experience with other human agents following those rules, thus potentially improving their acceptance of the robot's presence.

I was an author of two papers that appeared at HRI 2011:

E. Avrunin, J. Hart, A. Douglas, and B. Scassellati. Effects Related to Synchrony and Repertoire in Perceptions of Robot Dance. In Proceedings of the 6th International Conference on Human-Robot Interaction. March, 2011. 93-100. DOI

In this work we identify low-level aspects of robot motion that can be exploited to create impressions of agency and lifelikeness. In two experiments, participants view split-screen videos of multiple robots set to music and rate the robots on their dance ability, lifelikeness, and entertainment value. The first experiment tests the impact of the correspondence (or lack thereof) of the robot's motion to the underlying rhythm of the music, and the effect of matching changes in the robot's movement to changes in the music, such as a phrase of vocals or drumming. This motivates a second experiment which more deeply explores the relationships of asynchrony and changes in motion repertoire to participants' perceptions of the lifelikeness of the robot's motion. Findings indicate that perceptions of the lifelikeness of the robot and the quality of the dance can be manipulated by simple changes, such as variation in the repertoire of motions, coordination of changes in behavior with events in the music, and the addition of flaws to the robot's synchrony with the music.

D. Leyzberg, E. Avrunin, J. Liu, and B. Scassellati. Robots That Express Emotion Elicit Better Human Teaching. In Proceedings of the 6th International Conference on Human-Robot Interaction. March, 2011. 347-354. DOI

Does the emotional content of a robot's speech affect how people teach it? In this experiment, participants were asked to demonstrate several "dances" for a robot to learn. Participants moved their bodies in response to instructions displayed on a screen behind the robot. Meanwhile, the robot faced the participant and appeared to emulate the participant's movements. After each demonstration, the robot received an accuracy score and the participant chose whether or not to demonstrate that dance again. Regardless of the participant's input, however, the robot's dancing and the scores it received were arranged in advance and constant across all participants. The only variation between groups in this study was what the robot said in response to its scores. Participants saw one of three conditions: appropriate emotional responses, often-inappropriate emotional responses, or apathetic responses. Participants that taught the robot with appropriate emotional responses demonstrated the dances, on average, significantly more frequently and significantly more accurately than participants in the other two conditions.

In the summer of 2010, I participated in an REU run by the Quality of Life Technology Center at Carnegie Mellon University and the University of Pittsburgh. I worked with Professor Reid Simmons and one of his graduate students, Robert Fisher, on a system for preventing pressure ulcers in power wheelchair users by reminding them to adjust the posture of the chair at appropriate intervals. To make the reminders safe and effective, the behavior of the system must take into account the user's preferences and the current surroundings and situation, such as whether the chair is indoors or outdoors and what the user is doing. Our task was to create this context awareness from the data provided by sensors on the chair. My part in the project involved using accelerometer data to determine the surface the wheelchair is on, and I wrote code to evaluate how well different machine learning algorithms could do this terrain classification.

I have also worked on a project on kinematic self-modeling with a humanoid robot. We were developing a system to allow a robot to model the joints of its body by observing its own motion. This would let the robot learn the parameters and operation of its physical system without hardcoding, making possible on-line updates to the model as hardware is added or fails. In our system, the robot moves one joint and observes the location of its end effector. From a series of such motions, it calculates the parameters of the link. Single joints can then be chained together to create a complete model of the body. An extended abstract on this work appeared at HRI '09.

J.W. Hart, E.R. Avrunin, D. Golub, B. Scassellati, and S.W. Zucker. Incorporating active vision into the body schema. In Proceedings of the 4th ACM/IEEE International Conference on Human-Robot Interaction. March, 2009. 315-316. DOI

Department Activities

I have been an officer of RoboOrg, the graduate students' organization of the Robotics Institute, since April 2014.

I am a volunteer with Creative Technology Nights, an outreach program for middle school girls.

At Yale, I was a member of the Computer Science Department Student Advisory Committee, which represents undergraduate computer science majors and provides a liaison between the faculty and undergraduates. DSAC is elected annually by the Computer Science students.

I was also the treasurer and then president of Women Active in Computer Science at Yale, a student-founded organization to support and promote the participation of women in computer science.

Other Activities

For three years, I was the treasurer of the Yale Undergraduate Swing and Blues Dance Club. This is the non-competitive swing and blues dancing community of Yale and New Haven. We are dedicated to making social dancing available to every sector of our community by providing space for people to gather and dance in a fun, informal, welcoming setting, and by offering affordable opportunities for members of our community to learn.

I have a second degree black belt in Tae Kwon Do. I studied at Amherst Martial Arts for nine years, and was an instructor for two years.