24.04 10:30 - 11:30
USI East Campus, Room D0.10 |
|---|
| |
|---|
Abstract: Robots can be highly capable, but without understanding human preferences, their utility remains limited. This talk presents a framework for human augmentation through embodied social intelligence across three dimensions: human-centered design, embodied AI systems, and in-the-wild evaluation. I discuss methods to involve people throughout the robot lifecycle while balancing general-purpose systems with specific human needs, approaches to robust hardware and context-agnostic algorithms, and new evaluation metrics that bridge experimental rigor and real-world performance. Drawing from experience across academia and industry, I highlight open challenges at the intersection of AI, robotics, and design. Ultimately, I argue for a shift from intelligence to relationships, prioritizing human needs to build meaningful, adaptive, and accepted robots.
Chair: Prof. Monica Landoni | |
|---|
|
|---|
|
|
| | Patricia Alves-Oliveira is an Assistant Professor at the University of Michigan and Advisory Board member at Meta. She leads the Robot Studio, advancing human augmentation through embodied social intelligence, with a focus on healthcare. Her work spans human-centered robot design, embodied AI, and real-world evaluation, developing novel hardware and context-agnostic algorithms to integrate robots into everyday life. Previously, she was a Senior UX Designer at Amazon Lab126, contributing to Astro, and a postdoctoral researcher at the University of Washington. She holds a Ph.D. from the University Institute of Lisbon, with time at Cornell University. Her research has received three Best Paper Awards at ACM/IEEE HRI. She is co-founder of Talking Robotics and has been recognized as a Rising Woman Star in Social Robotics (2025), EECS Rising Star (2020), and Future DigiLeader (2020). 14:00 |
|---|
| |
|---|
|
|
|
|
