Fumin Pang
Papers
3
Total Citations
40
H-Index
3
About
Fumin Pang is a leading researcher in autonomous navigation, with a primary focus on visual-inertial odometry (VIO), sensor fusion, and simultaneous localization and mapping (SLAM) for ground robots. His most influential work addresses the critical challenge of scale observability in monocular VINS by tightly coupling visual-inertial data with wheel encoder measurements, introducing a novel wheel slip estimation model that significantly improves localization accuracy for wheeled robots. This work, his most cited with 24 citations, provides a practical solution for real-world deployment. Pang further advanced the field by developing a depth-enhanced VIO system based on the Multi-State Constraint Kalman Filter, demonstrating how sparse depth information can robustly improve state estimation. Beyond algorithmic contributions, he made a notable impact on the research community by co-creating the Segway DRIVE Benchmark, a comprehensive dataset collected from a fleet of delivery robots. This dataset directly addresses the critical gap between academic SLAM research and in-situ commercial operations, providing a realistic testbed for place recognition and SLAM algorithms in last-mile delivery scenarios. His work is essential for engineers developing reliable navigation systems for autonomous ground vehicles.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3