About Me
ABOUT THE PERSON BEHIND THE NOTES
Research is a way of
staying curious.
I am Zedong Yu (Yu Zetong), based in Shanghai. My work started with autonomous-driving perception and is now moving toward the larger question of how an intelligent system can understand, predict, and act in a changing world.
What I am connecting
Perception
Radar, camera, and 3D understanding for robust scene perception.
World models
Representations that make the future explicit enough to reason about.
Agents & VLMs
Multimodal systems that turn context into useful decisions and tools.
How this site works
The blog is for dated notes, experiments, and ideas in motion.
The knowledge base is for concepts that deserve a stable home.
The GitHub profile is where the code and open-source work live.
“The best note is not the one that sounds finished. It is the one that helps the next person take a better next step.”
Zedong Yu (于泽东)
- 📍 Shanghai
- 💼 AD perception → AD World model & LLM agents & VLM
- 🔗 GitHub
Background
I am a researcher focused on Autonomous Driving perception, now transitioning to:
- AD World Model
- LLM Agents
- Vision Language Models (VLM)