Jiaxu Xing
PhD student · Robotics and Perception Group, University of Zurich
I am a PhD student in the Robotics and Perception Group, under the supervision of Prof. Davide Scaramuzza. I am also an affiliated Doctoral student at the ETH AI Center.
My research focuses on robotics, computer vision, and machine learning.
Currently, I am a Research Intern at NVIDIA Research, working on world action models for manipulation.
Prior to this, I earned my MSc at ETH Zurich, where I had the privilege of working at the Autonomous Systems Lab, under the supervision of Prof. Roland Siegwart.
Research
My research interest lies in agile robot learning with strong generalization capabilities.
Specifically, I am working on developing high-performance, generalizable control policies for agile robots that can be effectively deployed in real-world environments.
News
| Oct 01, 2026 | Our paper Learning to Throw: Agile and Accurate Cable-Suspended Payload Delivery with a Quadrotor received the Best Poster Award at the IROS 2026 Sim2Real Workshop! |
|---|---|
| Sep 04, 2026 | Two papers are accepted to CoRL 2026! |
| Aug 03, 2026 | Excited to join NVIDIA Seattle Robotics Lab as a Research Intern! |
| Jul 13, 2026 | I am co-organizing the 8th Robot Learning Workshop: Is Physical AI Going Zero-Shot? at NeurIPS 2026 in Sydney! |
| Jun 18, 2026 | Our collaboration with RSL, CVG, and Princeton, What Matters in RL-Based Methods for Object-Goal Navigation? An Empirical Study and A Unified Framework, is accepted to ECCV 2026! |
| Jun 16, 2026 | Learning Acrobatic Flight from Preferences is accepted to IROS 2026! |
| May 15, 2026 | Learning Agile Quadrotor Flight in the Real World is accepted to RSS 2026! |
| Jan 31, 2026 | Sight Over Site is accepted to ICRA 2026! |
| Dec 17, 2025 | Learning on the Fly: Rapid Policy Adaptation via Differentiable Simulation is accepted to RA-L! |