Robots need to be able to respond to and learn from dynamic environments if they are going to spend extended amounts of time around people.
A 5-Year Collaboration Between 6 Universities and Industry Partners Will Place Assistive Robots in Everyday Human Environments.
A new National Science Foundation Science and Technology Center, led by the University of Texas at Austin, will study what happens when assistive robots leave the lab and move into the real world to live and work alongside humans.
The forthcoming Center for Human and Robot Co-Adaptation will bring together 39 researchers from six universities and multiple industry partners to establish a new science of “human-robot co-adaptation” — how people and robots learn from one another and adjust their behavior over long periods in everyday settings that will include homes, hospitals and assisted-living residences.
Daniel Brown
The University of Utah will play a central role in the new Center for Human and Robot Co-Adaptation. Led by Daniel Brown, PhD, assistant professor in the John and Marcia Price College of Engineering’s Kahlert School of Computing, the Utah contingent of researchers has expertise spanning human-AI alignment and safety, robot learning and manipulation, human-centered computing, and human factors.
“A lot of robotics research today focuses on whether a robot can successfully complete a particular task,” Brown said. “Through this Center, we want to dive deeper and understand what happens when a person and a robot live and work together for weeks, months, or even years.”
“This kind of long-term study of co-adaptation is really only possible with the kind of large-scale effort afforded by the Center,” he said. “I’m really excited about the opportunity to get the research we’ve been developing in the lab into hospitals, dormitories, museums, and care facilities to better understand how to create human-robot interactions that benefit and empower people.”
A patient uses sip-and-puff technology to play a video game, an example of patient-centered technology helping inform the future of assistive robotics.
The U will also provide one of the Center’s most important real-world proving grounds. Through the Center’s Human Environment with Robots (HERO) Facility Network, robots will be deployed at University of Utah Health sites, including the Craig H. Neilsen Rehabilitation Hospital and the planned One U Residential Center, designed to support independent living for people with complex physical injuries.
Craig H. Neilsen Rehabilitation Hospital is a leader in patient-facing technology and innovation, providing a clinical environment to develop and test new technologies that can increase patient independence, improve rehabilitation outcomes and support care teams. The project builds on a longstanding collaboration between Jason Wiese, PhD, associate professor in the Kahlert School of Computing; Jeffrey Rosenbluth, MD, Vice Chair of Innovation and Community Programs at Neilsen Rehabilitation Hospital; James Gardner, OTR/L, ATP, who coordinates the development of new patient-facing technology at Neilsen Rehabilitation Hospital; and Brent Eliesen, MBA, Health Innovation Team Lead with University of Utah Information Technology (UIT).
Their work includes the hospital’s 75 smart patient rooms, where assistive technologies are integrated into patient care and studied in a clinical setting. This established infrastructure and experience provide a strong foundation for expanding this work into assistive robotics. As a result of this innovative environment, Nielsen Rehabilitation Hospital will be the Center’s only hospital deployment site.
As part of the Center, fleets of mobile robots such as these XLE Robots from Dr. Kenneth Marino’s lab will be deployed in homes, hospitals, cafes, and museums to study how they can beneficially interact and co-adapt with humans. At the University of Utah, the robots shown here will be deployed in the Neilsen Rehabilitation Hospital to study how they can assist patients and staff members, starting with simple fetch-and-delivery tasks, but then expanding to a wider variety of tasks such as cleaning, tidying, social conversations, assistive feeding, and social navigation.
These real-world settings will allow researchers to study how patients, caregivers, healthcare professionals, and robots adapt to one another during extended, real-world interactions—moving robotics research beyond short laboratory experiments toward technologies that can safely support rehabilitation and independent living. Utah will also contribute to the Center’s education and outreach efforts through collaborations with the Utah STEM Action Center.
“This project gives us a chance to study robotics where the technology really matters: alongside people in their everyday environments,” Wiese said. “We want to understand how robots can adapt to individual needs and support greater independence.”
The U’s team will contribute across the Center’s major research areas:
Daniel Brown
Assistant Professor, Kahlert School of Computing; Utah Robotics Center
Human-AI alignment, robot learning, and human-robot interaction. He will work on aligning and inferring individual values and preferences, including how robots learn human preferences, values, constraints, and performance metrics.
Tucker Hermans
Professor, Kahlert School of Computing; Utah Robotics Center
Robot manipulation, planning, perception, and learning. He contributes to lifelong human-robot co-learning, robot competency/introspection, and helping robots learn and improve physical skills through long-term interaction.
Jason Wiese
Associate Professor, Kahlert School of Computing
Human-centered computing and human-robot interaction. His research includes smart/instrumented environments, accessibility, wellbeing, and technology for people with motor disabilities, particularly how technology fits diverse users and real-world contexts.
Kenneth Marino
Assistant Professor, Kahlert School of Computing; Utah Robotics Center
Semantic knowledge, NLP, computer vision, and AI. In the Center, Marino contributes to communication and shared cognitive/computational state—essentially helping robots reason about what people mean and establish common ground during interaction.
Frank Drews
Professor, Department of Psychology, College of Social and Behavioral Science
Behavior-centered safety and human factors. In the Center, Drews contributes to models of robot competency and trust, studying how people’s perceptions of robot capabilities relate to robots’ actual capabilities and how those perceptions evolve over time.
Other partner universities are Indiana University Bloomington, the Massachusetts Institute of Technology, Tufts University, and Yale University. Industry collaborators include Amazon, Apptronik, Diligent Robotics, Google DeepMind, Hello Robot, MassRobotics, NVIDIA and Robust AI.
Continue reading about the Center for Human and Robot Co-Adaptation at UT Austin.
