November 5
@
4:00 PM
–
5:00 PM
The ME 903: Graduate Student Lecture Series features campus and visiting speakers who present on a variety of research topics in the field of mechanical engineering. Professor Katie Fitzsimons is an Assistant Professor at Penn State.
Presentation Title: Learning from Variation: Adaptation and Personalization in Human–Robot Interaction
Abstract: Robots that interact with people must do more than execute precise motions: they must adapt to individual users, support learning, and respond meaningfully to human variability. This talk explores variation as both a mechanism for promoting neuromotor learning and a source of information for personalizing robot behavior.
The first part will focus on robotic systems for neuromotor rehabilitation. While traditional controllers often seek to minimize trajectory error, motor learning benefits from “repetition without repetition”—practicing tasks through varied movements and strategies. I will present adaptive controllers that encourage structured movement variability using probabilistic task representations and control objectives inspired by sensor-coverage problems. These methods enable robots to promote exploration while maintaining task relevance and user safety. The second part will examineshow robots can learn from human demonstrations and corrective feedback. Variation across demonstrations and corrections can reveal a user’s preferences, intent, and desired task strategy. I will describe approaches for using human feedback to adapt robot policies and personalize assistance. Together, these research directions view variation not as error to eliminate, but as a valuable resource for designing adaptive, personalized, and human-centered robotic systems.
Bio: Dr. Katie Fitzsimons is an Assistant Professor in the Department of Mechanical Engineering at The Pennsylvania State University. She received her Bachelor of Science degree in Mechanical Engineering from Michigan State University in 2013. She subsequently earned her Master of Science and Ph.D. degrees in Mechanical Engineering from Northwestern University in 2017 and 2020, respectively. Her research focuses on computational methods in human–machine interaction, robot learning from demonstration, and haptics. Her work integrates robotics, neuroscience, and health sciences to advance adaptive and human-centered robotic systems. Dr. Fitzsimons’ research applications include assistive exoskeleton control, information-based haptic feedback, and robot learning from multimodal human feedback, with the broader goal of improving collaboration between humans and intelligent machines. Her interdisciplinary work has been published in leading journals, including Science Robotics, ACM Transaction on Human-Robot Interaction, and IEEE Transactions on Haptics. She currently serves as an Associate Editor for IEEE Transactions on Medical Robotics and Bionics and ACM Transactions on Human-Robot Interaction.