I am a humanoid robotics researcher at the National Embodied Intelligence Robotics Innovation Center (Humanoid Innovation Department), supervised by Jony Zhang. I earned my M.Sc. in Human and Biological Robotics from Imperial College London and my B.A.Sc. in Mechanical Engineering with a minor in Robotics & Automation from the University of Toronto. Previously, I was a research intern at Tencent Robotics X, mentored by Shuai Wang. My research focuses on humanoid whole-body control, reinforcement learning for locomotion, motion retargeting, and Sim2Real deployment.
FlowOp uses sceneflow-conditioned diffusion to retarget arbitrary animation character motions to humanoid robots without body correspondence.
RobotPan predicts metric-scaled compact 3D Gaussians from sparse surround-view inputs for real-time 360° rendering and reconstruction on humanoid robots.
Heracles is a state-conditioned diffusion middleware that bridges precise motion tracking and generative synthesis for general-purpose humanoid control.
MeshMimic bridges 3D scene reconstruction and embodied intelligence to enable humanoid robots to learn coupled motion-terrain interactions directly from monocular video.
TAIROS is an open embodied-AI platform that provides modular large models, cloud simulation, data tooling, and SDKs to support the full robotics development lifecycle from training to real-world deployment.
* denotes equal contribution, † denotes corresponding author.
A wheel-leg humanoid research prototype for domestic environments, featuring switchable wheeled/legged locomotion, tactile skin, and dexterous hands for elder-care tasks such as assisted mobility and object fetching.
A wheel-legged robot that combines efficient wheeled mobility with legged obstacle crossing, dynamic balance control, and tactile sensing for agile locomotion on flat ground and complex terrain.
Approximate visitor locations based on IP geolocation. Click the map for full statistics.
