Onur Bagoren
Papers
7
Total Citations
57
H-Index
4
About
Onur Bagoren is an emerging robotics researcher specializing in underwater robot perception, simultaneous localization and mapping (SLAM), and machine learning for maritime applications. His work addresses some of the most demanding challenges in autonomous underwater vehicles (AUVs), particularly operating in low-texture, acoustically complex, and highly unstructured environments where conventional methods routinely fail. Bagoren's most impactful contribution is his development of an open-source benchmark dataset for shipwreck segmentation from side scan sonar imagery, which has already garnered 28 citations and provides the underwater robotics community with a critical resource for training and validating machine learning models. His TURTLMap system advances real-time dense mapping using low-cost AUVs, while his uncertainty-aware acoustic localization framework addresses the fundamental challenge of unreliable sensing in dynamic underwater conditions. More recently, he has pushed boundaries with OceanSim, a GPU-accelerated underwater perception simulation framework, and SonarSplat, an innovative application of Gaussian splatting to imaging sonar for realistic novel view synthesis. Collectively accumulating over 57 citations across recent publications, Bagoren's research is rapidly shaping the infrastructure—datasets, simulators, and perception algorithms—that will underpin the next generation of intelligent underwater robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6SonarSplat: Novel View Synthesis of Imaging Sonar via Gaussian Splatting2 citations · 2025
- 7