Papers
26
Total Citations
3,752
H-Index
19
About
Michael Bosse is a prominent robotics researcher whose work has profoundly shaped the fields of simultaneous localization and mapping (SLAM), visual-inertial odometry, and mobile 3D perception. His early contributions through the Atlas framework — a scalable, hybrid metrical/topological approach to SLAM — established foundational methods for mapping large-scale cyclic environments, accumulating over 600 citations across two landmark papers. His collaborative work on keyframe-based visual-inertial odometry using nonlinear optimization has become a cornerstone reference in mobile robotics, amassing nearly 1,700 citations and influencing countless autonomous navigation systems. Bosse also pioneered innovative sensor design, most notably the Zebedee spring-mounted 3D range sensor, which enabled flexible and practical mobile mapping solutions (400 citations). His research extends into lifelong localization, where he developed "summary map" techniques that allow robots to maintain robust performance across changing environments over extended deployments. Additional contributions span terrain assessment for ground vehicles, continuous-time aerial mapping, and 2D lidar place recognition. Together, his body of work reflects a career dedicated to making autonomous robots reliably perceive, map, and navigate complex real-world environments at scale.
Research Focus
Key Achievements
Top Papers
- 1Keyframe-based visual–inertial odometry using nonlinear optimization1,697 citations · 2014
- 2
- 3An Atlas framework for scalable mapping313 citations · 2003
- 4
- 5
- 6Keypoint design and evaluation for place recognition in 2D lidar maps124 citations · 2009
- 7Summary Maps for Lifelong Visual Localization108 citations · 2015
- 8Mapping Partially Observable Features from Multiple Uncertain Vantage Points83 citations · 2002
- 9
- 10The gist of maps - summarizing experience for lifelong localization76 citations · 2015