Ivan Petrovi
Papers
2
Total Citations
11
H-Index
2
About
Ivan Petrovi is a researcher whose work sits at the intersection of robotics, sensor fusion, and probabilistic state estimation. His primary contributions lie in advancing simultaneous localization and mapping (SLAM) for mobile robots operating under challenging conditions, as well as developing robust probabilistic models for acoustic sensor arrays. In his highly cited 2010 paper, "Exactly Sparse Delayed State Filter based robust SLAM with Stereo Vision" (6 citations), Petrovi introduced a novel SLAM framework that leverages a constrained stereo vision image registration algorithm, designed to remain robust even in the presence of high odometry uncertainty. This work directly addresses a critical bottleneck in real-world autonomous navigation. Complementing this, his 2010 paper "Applying von Mises Distribution to Microphone Array Probabilistic Sensor Modelling" (5 citations) demonstrates his versatility by applying circular statistics to the problem of full-range speaker localization and tracking. By modeling time difference of arrival with a von Mises distribution, Petrovi enabled a mobile robot to probabilistically track a moving speaker across 360 degrees. These contributions showcase his ability to fuse geometric and probabilistic reasoning, producing algorithms that are both theoretically sound and practically deployable for robust robotic perception.
Research Focus
Key Achievements
Top Papers
- 1Exactly Sparse Delayed State Filter based robust SLAM with Stereo Vision6 citations · 2010
- 2