Nadia Figueroa, the Shalini and Rajeev Misra Presidential Assistant Professor in Mechanical Engineering and Applied Mechanics (MEAM) in Penn Engineering, has received the National Science Foundation’s Faculty Early Career Development (CAREER) Award, one of the most prestigious honors supporting early-career faculty whose work integrates research and education in meaningful and far-reaching ways.

Her award-winning project, Constraint-Aware Estimation, Learning and Control for Fluid Physical Human-Robot Collaboration, aims to fundamentally expand how robots sense, predict and physically interact with people in dynamic, real-world environments.

“We want robots to be truly useful, physically intelligent machines that understand us, adapt to us, and keep us safe,” Dr. Figueroa said. “To get there, we need robots that can interact with humans as fluidly as humans interact with one another.”

In her lab, Dr. Figueroa and her students work across control, learning, perception, and robot design to achieve “fluid physical interaction,” a regime where robots co-regulate motion, force and intent continuously, adapting to uncertainty just as humans do.

“What excites me most is building robots that can understand the world the moment they touch it,” Dr. Figueroa said. “Our algorithms bring sensing, learning, and control together so a robot can instantly judge how heavy something is, adjust its strength, or follow a person’s guidance through touch. I want to build robots that can help lift or carry objects safely, open jars or doors, polish a surface with human feedback, or handle items it has never seen before. And with a new humanoid robot coming to the lab next year, we’ll be able to explore even more complex, whole-body tasks like carrying boxes, navigating cluttered spaces, and truly working alongside people in a natural way.”

Dr. Figueroa collaborates closely with her campus colleagues Josh Baxter in Penn Medicine and Flavia Vitale in Penn Engineering and Penn Medicine to understand human biomechanics and apply those principles to robot design.

One project, a “muscle-in-the-loop” training platform, allows robotic arms to adjust their resistance based on real-time measurements of a participant’s muscle tension and strength. The system has applications in rehabilitation, athletics, and the study of human motor adaptation, and may ultimately inform muscle-inspired robot controllers.