Haodong Zhu 朱昊东

I am an undergraduate student in the Department of Automation, Tsinghua University. Since April 2025, I have been a research intern at the Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University, advised by Prof. Yang Gao. My work explores how demonstration data, vision-language-action models, and whole-body control can enable humanoid robots to perform complex mobile manipulation tasks in real-world environments.

Research Interests

My research interests center on embodied intelligence and humanoid loco-manipulation. In the long term, I aim to build a general-purpose robot brain that can understand language and vision, reason about the physical world, anticipate the consequences of its actions, and transfer knowledge across robot embodiments, tasks, and environments.

Email  /  Scholar  /  Github  /  X  /  WeChat

Publications (* indicates equal contribution)

We present OpenHLM, an empirical recipe for naive whole-body humanoid loco-manipulation 🦿👐. With controlled studies on teleoperation, VLA design, and heterogeneous co-training, OpenHLM tackles diverse language-conditioned tasks 📕 — and even goes fully autonomous in the wild 🌳, no mocap required.

We present HuMI (Humanoid Manipulation Interface), the first robot-free 🚫🤖 framework for learning diverse humanoid whole-body manipulation tasks across various environments 🏠.

Modified from Jon Barron's website.