I'm interested in making robots adaptive and safe enough to stay reliable in real human environments, not just in the lab. I believe that requires tight integration of perception, decision making, and control, leveraging model-based structure and learned adaptation. In my master's thesis, I'm working toward risk-aware decision making and control for humanoids, building on my research in uncertainty-aware, self-supervised proprioceptive state estimation.
I'm an M.S. in Robotics student and Rales Fellow in the AirLab at CMU, working with Sebastian Scherer on state estimation and humanoid planning and control. Before this, I studied Electrical Engineering at Georgia Tech. There I worked in Patricio Vela's IVALab on closed-loop SLAM benchmarking and on an RL agent that tunes navigation and SLAM parameters online, and in Christopher Rozell's SIPLab, where I led a team learning representations that separate natural neural dynamics from stimulation-driven effects for use in control. Along the way I've also worked on embedded systems and FPGA firmware at Sandia National Labs and Southwest Research Institute.
Timeline
-
FellowshipInducted as a CMU Rales Fellow
-
JoinedStarted the M.S. in Robotics at Carnegie Mellon, continuing in the AirLab
-
GraduatedB.S. Electrical Engineering, minor in Computer Science, Georgia Tech
Provost's Academic Excellence Award (College of Engineering) · HKN Outstanding EE Senior · College of Engineering Outstanding Undergraduate Researcher
-
ResearchJoined the CMU AirLab, working on learned inertial state estimation for legged robots
-
IROSTask-driven SLAM Benchmarking for Robot Navigation
Y. Du, S. Feng, C. Cort, P. A. Vela · IROS 2025
-
First paperLow-Cost Real-World Benchmark Framework for Closed-Loop Robot Navigation Systems with SLAM
C. Cort, A. Dalal, J. De Ocampo, Y. Du, S. Feng, P. A. Vela · IEEE Opportunity Research Scholars Symposium (ORSS) 2024