Marius Fehr

ETH Zurich

Papers

16

Total Citations

693

H-Index

11

About

Marius Fehr is a robotics researcher specializing in visual-inertial mapping, autonomous exploration, and 3D scene understanding — areas at the forefront of modern mobile robotics. He is perhaps best known for his foundational contributions to **maplab**, an open-source framework for visual-inertial mapping and localization that has become a widely adopted research tool, accumulating over 270 citations since its 2018 release. Its successor, maplab 2.0, extended this work with modular, multi-modal SLAM capabilities, reflecting Fehr's sustained commitment to robust, scalable mapping systems. Fehr has also made significant contributions to 3D reconstruction, introducing TSDF-based change detection methods for long-term dynamic environments and contributing to the Voxblox planning framework. His work on hybrid topological-dense mapping addresses the real-world challenge of operating in large indoor spaces efficiently. A notable applied achievement is his involvement with the **CERBERUS** team in the DARPA Subterranean Challenge, where autonomous legged and aerial robots demonstrated complex underground exploration — work that has garnered over 100 citations. With a portfolio spanning sensor calibration, topological navigation, and open hardware design, Fehr's research consistently bridges theoretical innovation with practical robotic deployment, making him a valuable reference for students working in SLAM, autonomous navigation, and field robotics.

Research Focus

Key Achievements

11
H-Index
16
Papers
693
Total Citations
43
Avg Citations/Paper
🏆 Most Cited Paper
Maplab: An Open Framework for Research in Visual-Inertial Mapping and Localization
270 citations · 2018
📈 Most Prolific Year: 2018 (5 Papers)
🤝 Key Collaborators: 76
🏛 Institutions: ETH Zurich

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago