Xingyu Chen

Currently, I am a PhD student focus on humanoid, and start my research journey with Prof. Renjing Xu in HKUST(GZ).

Previously, I am a Robotics Animation Engineer at Xiaomi in 2025. Prior to that, I obtained M.Sc degree from University College London in 2024.

Email  /  Resume  /  Github  /  Scholar  /  LinkedIn

Highlight

[Open] Welcome research collaborations on humanoid technique
I co-founded a cross-institute research group (member from 清华,上交,复旦,浙大,小米,etcs) in following topics:

  • Humanoid whole-body control
  • Interaction skill with object/scene
  • Multi-modal motion representation

If you aims to solve real-world problem and extend the tech boundary on humanoid, feel free to reach me out, some positions are avaiable in both university and industry, I'm glad to recommend you involve in and offer assistance to support you.

Research

Particularly, I am interested in physical intelligence via spatial computing.

HAIC: Humanoid Agile Object Interaction Control via Dynamics-Aware World Model
Dongting Li*, Xingyu Chen*, Qianyang Wu, Bo Chen, Sikai Wu, Hanyu Wu, Guoyao Zhang, Liang Li, Mingliang Zhou, Diyun Xiang, Jianzhu Ma, Qiang Zhang, Renjing Xu
Robotics: Science and Systems (RSS), 2026
[project page] [arXiv] [code]

A world model completes the full-state object observation during dynamic interaction.

PDF-HR: Pose Distance Fields for Humanoid Robots
Yi Gu*, Yukang Gao*, Yangchen Zhou*, Xingyu Chen, Yixiao Feng, Mingle Zhao, Yunyang Mo, Zhaorui Wang, Lixin Xu, Renjing Xu
Arxiv preprint, 2026
[project page] [arXiv] [code]

A continuous distance field for modeling humanoid robotics pose distribution.

Implicit Kinodynamic Motion Retargeting for Human-to-humanoid Imitation Learning
Xingyu Chen, Hanyu Wu, Sikai Wu, Mingliang Zhou, Diyun Xiang, Haodong Zhang
Arxiv preprint, 2025
[project page] [arXiv] [code]

A real-time large-scale human-to-humanoid motion retargeting by style transfer.

Text-driven Human Motion Generation with Motion Masked Diffusion Model
Xingyu Chen
Arxiv preprint, 2024
[arXiv]

A spatial-temporal mask strategy for text-driven human motion diffusion model.

Gaussian Surfel Representation for Visual SLAM
Xingyu Chen
Dissertation for master degree
[thesis]

A master thesis on the Visual SLAM with a planer Gaussian surfel representation.

Engineering

I also enjoy some system design, implementation, product deployment and engineering applications.

Sensor Fusion and Dynamic Simulation for Autonomous Driving
Internship guided by Dr.Panqu Wang
ZERON, 2024

I led and responsible for the 3DGS closed-loop simulation of the E2E autodrive model.

Robotic Sensing, Manipulation and Interaction
Lecture instructed by Dr.Francisco
UCL COMP0129, 2023
[github]

A point cloud recognition, positioning and grasping for Franka Panda Arm in Gazebo.

VR Game for Multi-player Online Helicopter Interaction
Project developed based on Ubiq
UCL COMP0113, 2023
[github]

A multiplayer real-time synchronous game for helicopter driving in Unity on Meta Quest2.

Mobile robotics clusters for industrial routing inspection
Internship guided by Dr.Jingya Liu
MCC CISDI, 2021

I was involved and responsible for the SLAM and navigation system in collaboration robot.

Visual servo robotics for metallurgical manufacturing industry
Internship guided by Dr.Jingya Liu
MCC CISDI, 2020

I was involved and responsible for the simulation and integrated control in industry robot.

Miscellanea
Waterbot's Notebook

If you are interested in the humanoid whole-body control, I maintain a blog to help beginners quickly go through key points and getting started.

[My mkdocs]

This web template is from Jon Barron, the head icon is from 小熊貓, thank all!