Yifan Hou

I am a Postdoctoral researcher at the REALab, Stanford University, working with Prof. Shuran Song. Prior to joining Stanford, I was an Applied Scientist on the Vulcan team at Amazon Robotics. I obtained my PhD and MS degrees from the Robotics Institute at Carnegie Mellon University, advised by Prof. Matthew T. Mason. I obtained BoE from the department of Automation at Tsinghua University. I had also spent time at Toyota Research Institute and MIT.
My career goal is to achieve human level manipulation dexterity on robots. During my PhD, I designed methods to execute contact-rich manipulation robustly against modeling uncertainties and disturbance forces using active compliance control schemes. I then spent three years at Amazon Robotics pushing compliant manipulation from research to industrial products in the Stow project. I believe that robotic manipulation has an opportunity of wide adoption in people’s daily life, where the bottleneck is the ability to scale up the aquisition of robust, general manipulation skills. Currently my research interests include the intersection of data-driven visual motor policies and robust model-based control as well as dexterous manipulation.
news
Jun 03, 2025 | DexMachina is released! We propose a solution to learn dexterous hand manipulation from a single manipulation. |
---|---|
May 30, 2025 | DexUMI is released! We propose a dexterous hand data collection system that bridges the embodiment gap. |
May 25, 2025 | I gave an invited talk “Empower Robot Learning with Model-based Manipulation” at the Beyond Pick and Place workshop at ICRA 2025. |
May 21, 2025 | Adaptive Compliance Policy won the best paper award at the Contact-rich Manipulation workshop at ICRA 2025. |
Feb 19, 2025 | I recently gave a lecture on “Introduction to Compliance Control” at Stanford EE/CS 381. Slides are available here. |
Jan 27, 2025 | Adaptive Compliance Policy is accepted to ICRA 2025. See you in Atlanta! |
Jan 17, 2025 | My invited talk at the Stanford Robotics Seminar was released. This talk is an overview of my work on compliant manipulation. |
Oct 14, 2024 | Our recent work on Adaptive Compliance Policy is released! |
Jun 26, 2024 | Our paper simPLE: a visuotactile method learned in simulation to precisely pick, localize, regrasp, and place objects has been published at Science Robotics. |
Mar 18, 2024 | I joined Stanford University as a Postdoctoral researcher. |