Hongjie Fang

About

Hongjie Fang is a leading researcher in robotic manipulation, whose work is reshaping how robots perceive, grasp, and interact with the physical world. His core research spans grasp perception, reactive manipulation, and large-scale robotic learning. Fang’s seminal contribution, **AnyGrasp** (210 citations), established a robust and temporally continuous grasp perception system, enabling robots to match human-like dexterity across space and time. He also pioneered **Graspness Discovery** (124 citations), a method that dramatically accelerates grasp detection by identifying where to grasp in cluttered scenes. Fang played a key role in the **Open X-Embodiment** collaboration (119+ citations), a landmark project that aggregated diverse robotic datasets to train generalist foundation models. His work on **TransCG** (116 citations) addressed the challenging problem of grasping transparent objects by introducing a large-scale depth completion dataset. More recently, Fang introduced **AirExo** (30 citations), low-cost exoskeletons for learning whole-arm manipulation, and **FoAR** (2025), a force-aware reactive policy for contact-rich tasks. With over 780 total citations and a trajectory from foundational grasp algorithms to cross-embodiment learning, Fang is shaping the next generation of general-purpose robotic manipulation.

Research Focus

Key Achievements

Top Papers

1. 1
   AnyGrasp: Robust and Efficient Grasp Perception in Spatial and Temporal Domains

   210 citations · 2023
2. 2
   Graspness Discovery in Clutters for Fast and Accurate Grasp Detection

   124 citations · 2021
3. 3
   Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration⁰

   119 citations · 2024
4. 4
   TransCG: A Large-Scale Real-World Dataset for Transparent Object Depth Completion and a Grasping Baseline

   116 citations · 2022
5. 5
   Open X-Embodiment: Robotic Learning Datasets and RT-X Models

   101 citations · 2023
6. 6
   RH20T: A Comprehensive Robotic Dataset for Learning Diverse Skills in One-Shot

   55 citations · 2024
7. 7
   AirExo: Low-Cost Exoskeletons for Learning Whole-Arm Manipulation in the Wild

   30 citations · 2024
8. 8
   Target-referenced Reactive Grasping for Dynamic Objects

   16 citations · 2023
9. 9
   Flexible Handover with Real-Time Robust Dynamic Grasp Trajectory Generation

   9 citations · 2023
10. 10
    FoAR: Force-Aware Reactive Policy for Contact-Rich Robotic Manipulation

    3 citations · 2025

Key Collaborators

Contact & Links