Wenqiang Xu
Papers
19
Total Citations
271
H-Index
8
About
Wenqiang Xu is a robotics and embodied AI researcher whose work bridges perception, simulation, and manipulation to advance intelligent robotic systems. His research spans articulated object understanding, robot learning, caregiving robotics, and tactile sensing, with a particular focus on closing the gap between synthetic training environments and real-world deployment. Xu's most influential contribution is AKB-48, a real-world articulated object knowledge base that has reshaped how researchers approach object structure and pose estimation, garnering 64 citations since 2022. Complementing this, his work on category-level articulation pose estimation pushes beyond single-instance constraints toward more generalizable solutions. His development of RCareWorld, a human-centric caregiving simulation platform, demonstrates a commitment to socially impactful robotics, while RFUniverse extends this vision to multiphysics simulation for embodied AI. Xu has also made notable contributions in tactile sensing, proposing a deep learning-powered stretchable array for capturing forceful interactions with deformable objects, and in transparent object manipulation through refractive flow estimation. Across his portfolio, Xu consistently delivers benchmark datasets, simulation platforms, and algorithmic frameworks that equip the broader robotics community with practical, generalizable tools for tackling real-world complexity.
Research Focus
Key Achievements
Top Papers
- 1AKB-48: A Real-World Articulated Object Knowledge Base64 citations · 2022
- 2Toward Real-World Category-Level Articulation Pose Estimation37 citations · 2022
- 3RCare World: A Human-centric Simulation World for Caregiving Robots35 citations · 2022
- 4
- 5H2O: A Benchmark for Visual Human-human Object Handover Analysis23 citations · 2021
- 6
- 7Demonstrating RFUniverse: A Multiphysics Simulation Platform for Embodied AI16 citations · 2023
- 8
- 9
- 10