Nolan Fey

I'm a Ph.D. student in the Improbable AI Lab at MIT Computer Science and Artificial Intelligence Lab (CSAIL), where I'm advised by Pulkit Agrawal. My research focuses on giving robots the ability to intelligently move through and interact with the world. I'm grateful to be supported by the NSF GRFP.

I'm also a Founding Research Lead at Minerva Humanoids, where we're deploying humanoid robots for the most dangerous jobs on the planet.

Previously, I received a B.S. in Electrical Engineering with a Second Major in Physics at Notre Dame, where I worked with Patrick Wensing in the ROAM Lab and with Michael Lemmon. I've also been fortunate to intern at the U.S. Naval Research Lab in D.C. and NASA JPL.

Email  /  Google Scholar  /  X  /  GitHub

Research

I equip robots with foundational motion skills. Recently I've been especially interested in sim-to-real transfer: how we can ensure that behaviors learned in simulation will work in the real world.

Agile humanoid dribbling, shooting, passing, and receiving with onboard active vision, enabled by data-driven alignment of perception and dynamics for sim-to-real transfer.

Aligning contact forces between hardware and simulation enables residual world modeling for floating-base humanoids, improving transfer of force interactions from just 6.5 minutes of walking data.

Variable-stiffness humanoid motion imitation enabling gentle contact with the environment for safety and generalization.

Enhancing the sim-to-real transfer for a legged manipulator by calibrating actuator dynamics and refining whole-body control, enabling agile feats like throwing, lifting, and dragging

A learning-based framework that allows a walking robot to stabilize itself under disturbances neglected by its base controller. We applied it to stabilize the MIT Mini Cheetah as it carried a box of water on its back.

By approximating each of the robot's stance phases to be impulsive, we enable a mixed-integer program to quickly plan consecutive hops for a quadruped between surfaces while avoiding obstacles.