Research

Selected research on human-centered embodied intelligence, humanoid self-representation, reliable embodied-agent systems, 3D vision, and interpretable scientific machine learning.

My current research develops foundations for human-centered embodied intelligence across humanoid self-representation, reliable embodied-agent systems, and humanoid motion and manipulation. A central direction is Thea & Unitree G1, which combines real-robot locomotion and manipulation with closed-loop tool use, persistent scene state, and independent action verification.

project thumbnail

Thea & Unitree G1

Deploying Unitree G1 locomotion and manipulation within a closed-loop embodied-agent harness for long-horizon tasks and failure recovery.
project thumbnail

Humanoid Self-Representation

Learning which body is the self—and how that body occupies 3D space—from sensorimotor experience.
project thumbnail

Interpretable Scientific ML with KAN

Physics-guided Kolmogorov–Arnold Networks that turn learned predictors into inspectable retrieval formulas.
project thumbnail

Generative 3D Rigging

Learning skeletons, skinning, and semantic correspondence for controllable 3D assets.