My research focuses on embodied AI and robot learning: building systems that perceive, act, and improve through interaction with the physical world. I am particularly interested in learning-based task and motion planning—including long-horizon skill composition, reasoning under partial observability, and integrating learned policies with structured planners for reliability.

I also care about generalization and deployment: transferring behaviors across embodiments and environments, grounding high-level goals in executable controllers, and closing the sim-to-real gap for fieldable autonomy in assistive, construction, or mobile-manipulation settings.