Michael Mashner

Papers

Total Citations

H-Index

About

Michael Mashner is a leading figure in probabilistic robotics, with a focus on the mathematical foundations of state estimation and sensor fusion. His research centers on applying differential geometry and Lie group theory to solve complex problems in robot localization and mapping. Mashner’s most influential work, “Bayesian Fusion on Lie Groups” (2011, 52 citations), provides a rigorous framework for generalizing classical Bayesian estimation to non-Euclidean spaces, enabling more accurate uncertainty representation in real-world robotic systems. This theoretical contribution has become essential for researchers working on 3D pose estimation and visual-inertial navigation. In his highly cited follow-up, “The Banana Distribution is Gaussian: A Localization Study with Exponential Coordinates” (2012, 45 citations), Mashner challenges conventional assumptions by demonstrating that seemingly non-Gaussian pose distributions—like the characteristic “banana-shaped” uncertainty of wheeled robots—can be elegantly modeled as Gaussian in exponential coordinates. This insight has improved the efficiency and accuracy of planar mobile robot localization algorithms. Mashner’s work bridges pure mathematics and practical robotics, offering tools that are both theoretically sound and computationally tractable, making him a key reference for students and researchers in probabilistic robotics and sensor fusion.