Benjamin Metka

Frankfurt University of Applied Sciences, Goethe University Frankfurt

Papers

Total Citations

H-Index

About

Benjamin Metka is a researcher in autonomous mobile robotics, with a primary focus on biologically inspired learning algorithms for spatial navigation and self-localization. His work centers on Slow Feature Analysis (SFA), a computational principle derived from the visual cortex that extracts slowly varying features from rapidly changing sensory input. Metka’s major contribution is demonstrating that hierarchical SFA can enable a robot to learn a spatial representation of its environment directly from raw camera images captured during an unguided random walk, without any external labels or prior maps. Notably, his 2017 paper on "Efficient navigation using slow feature gradients" shows that after unsupervised learning, a subset of these representations become orientation-invariant and effectively code for the robot’s position, allowing for robust localization. His 2013 paper on outdoor self-localization using SFA (10 citations) and his 2016 work on improving robustness through loop closure events (5 citations) further solidify his impact. Though his citation counts are modest, Metka’s research is pioneering in bridging unsupervised learning and spatial cognition, offering a path toward more autonomous and adaptive robotic navigation systems.