Mevlana Gemici

Papers

Total Citations

153

H-Index

About

Mevlana Gemici is a researcher whose work sits at the intersection of robotics, machine learning, and geometry. His key research areas include deformable object manipulation, density estimation on non-Euclidean spaces, and low-power autonomous navigation for aerial robots. Gemici’s most cited work (81 citations) tackles the challenging problem of teaching robots to manipulate deformable food objects by learning haptic representations from tool-based interactions—a critical skill for personal robotics. He also made foundational contributions to probabilistic machine learning by developing normalizing flows on Riemannian manifolds (46 citations), enabling density estimation on curved spaces essential for applications in robotics, protein folding, and physics. Earlier in his career, Gemici demonstrated innovation in embedded autonomy by creating low-power parallel algorithms for single-image obstacle avoidance in aerial robots (26 citations). His work bridges theoretical advances in geometric deep learning with practical robotic systems, showing how robots can perceive and interact with complex, unstructured environments.