Zi-ang Cao

I'm a first-year PhD student in UT Austin Robot Perception and Learning Lab (RPL), advised by Prof. Yuke Zhu. Previously, during my M.S. at Stanford, I had the privilege of working with Prof. Karen Liu in The Movement Lab and Prof. Jeannette Bohg in the Interactive Perception and Robot Learning Lab.

Collaborators: If you're interested in collaborating or learning more about my work, feel free to reach out to me via email and I am always happy to chat!

Email  /  Google Scholar  /  Github  /  Linkedin  /  Twitter

I am eager to delve deeper into robotic learning and cross-embodiment learning.

2025/08 - Two papers got accepted to CoRL 2025
2024/11 - I provided the demo House Heroes: The Stanford Tidy Taskforce at Stanford Robotic Center Launch Ceremony. | Please checkout our Twitter highlight to see a team of smart mobile robots empowered by robot learning to do chores in an apartment setup.
2024/08 - Two papers got accepted to CoRL 2024
2023/08 - One papers got accepted to CoRL 2023

Hand-Eye Autonomous Delivery (HEAD): Learning Humanoid Navigation, Locomotion and Reaching

Teach humanoid robots to navigate complex environments autonomously and deliver their hands to any specified target.

TWIST: Teleoperated Whole-Body Imitation System

Versatile, Coordinated, Whole-Body Skills for Humanoid Robots via Teleoperation.

Equi-Bot: Equivariance in Imitation Learning for Robust Mobile Manipulation

Exploring equivariant representations for enhanced generalization in robot imitation learning. Targeting mobile manipulation applications.

Unpacking Failure Modes of Generative Policies: Runtime Monitoring of Consistency and Progress

Robot behavior policies trained via imitation learning are prone to failure under conditions that deviate from their training data. In this work, we present Sentinel, a runtime monitor that detects unknown failures (requiring no data of failures) of generative robot policies at deployment time.

What Went Wrong? Closing the Sim-to-Real Gap via Differentiable Causal Discovery