Papers

6

Total Citations

448

H-Index

5

About

Yafei Hu is a robotics and artificial intelligence researcher whose work spans robot navigation, autonomous exploration, visual SLAM, and the integration of foundation models into general-purpose robotic systems. He is perhaps best known for his contributions to TartanAir, a landmark dataset designed to push the boundaries of visual simultaneous localization and mapping (SLAM). Collected within photo-realistic simulation environments featuring dynamic objects, shifting lighting, and varied weather conditions, TartanAir has become a foundational benchmark in the field, accumulating nearly 400 citations and demonstrating the power of simulation-driven data collection for training robust navigation systems. Beyond dataset development, Hu has made significant strides in autonomous exploration, proposing off-policy evaluation methods with online adaptation that enable robots to reason about future state values in challenging environments. His research on unsupervised robotic interestingness detection addresses the critical real-world constraint of limited training data, allowing robots to identify salient scenes with minimal supervision. More recently, Hu has explored the frontier of generalizable feature fields for mobile manipulation and contributed a comprehensive survey on leveraging foundation models for general-purpose robotics — reflecting his broad vision of creating adaptable, intelligent robotic systems capable of operating seamlessly across diverse real-world scenarios.

Research Focus

Key Achievements

5
H-Index
6
Papers
448
Total Citations
75
Avg Citations/Paper
🏆 Most Cited Paper
TartanAir: A Dataset to Push the Limits of Visual SLAM
365 citations · 2020
📈 Most Prolific Year: 2020 (2 Papers)
🤝 Key Collaborators: 35
🏛 Institutions: Carnegie Mellon University, UC San Diego Health System

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago