Papers
9
Total Citations
181
H-Index
7
About
Fei Wen is a leading researcher in autonomous navigation and robotics, specializing in multi-sensor fusion, visual simultaneous localization and mapping (V-SLAM), and robust perception systems. His major contributions include developing adaptive fusion frameworks that integrate GNSS with visual-inertial odometry to maintain accurate global positioning even under intermittent satellite degradation—a critical challenge for autonomous vehicles. Wen also pioneered a tightly coupled 10-degree-of-freedom optimization on manifold for GNSS and vision SLAM, addressing drift and scale uncertainty in relative positioning. His work on efficient maximum consensus robust fitting algorithms provides deterministic solutions for computer vision applications, while his point cloud recognition networks enhance rotation robustness for industrial robotics. With over 180 citations across his top papers, Wen’s research has significantly advanced SLAM-based topological mapping and navigation, integrating deep reinforcement learning for local planning. Notable achievements include his active SLAM methods that enable autonomous exploration from uncertain starting positions, and his recent BEVGM approach for visual place recognition using bird’s eye view graph matching to handle challenging appearance variations. Wen’s innovations directly impact autonomous driving, robotics, and industrial automation.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Efficient Algorithms for Maximum Consensus Robust Fitting20 citations · 2019
- 4
- 5SLAM Based Topological Mapping and Navigation14 citations · 2020
- 6
- 7Active SLAM With Prior Topo-Metric Graph Starting At Uncertain Position9 citations · 2021
- 8Direct-ORB-SLAM: Direct Monocular ORB-SLAM5 citations · 2019
- 9