Papers
8
Total Citations
93
H-Index
5
About
Duc Fehr is a roboticist whose research bridges perception, locomotion, and autonomous navigation. His most influential work focuses on 3D object recognition, where he pioneered the use of compact covariance descriptors for point clouds—a contribution that earned 38 citations and remains foundational for mobile robots interpreting their environment. Fehr extended this approach to RGB-D data, introducing novel feature descriptors that enable robots to classify objects using color and depth information simultaneously. Beyond perception, he has explored unconventional locomotion with tumbling robots, formalizing motion primitives for a novel class of highly mobile machines. His practical contributions include methods for object reconstruction using laser range-finders and strategies for camera placement in surveillance systems. Fehr also addressed the critical challenge of occlusion in robotic vision, demonstrating how mobile robots can actively reposition to overcome visual obstructions. With a citation count exceeding 90 across his key papers, Fehr’s work has influenced both theoretical frameworks in covariance-based descriptors and practical implementations in field robotics. His research at the University of Minnesota’s Center for Distributed Robotics reflects a consistent focus on enabling robots to see, understand, and move through complex environments.
Research Focus
Key Achievements
Top Papers
- 1Compact covariance descriptors in 3D point clouds for object recognition38 citations · 2012
- 2Issues and solutions in surveillance camera placement16 citations · 2009
- 3RGB-D object classification using covariance descriptors14 citations · 2014
- 4Motion primitives for a tumbling robot7 citations · 2008
- 5
- 6Occlusion alleviation through motion using a mobile robot5 citations · 2014
- 7
- 8