Matheus M. dos Santos
Universidade Federal do Rio Grande, Robotics Research (United States)
Papers
13
Total Citations
137
H-Index
6
About
Matheus M. dos Santos is a researcher specializing in autonomous underwater robotics, with a particular focus on acoustic perception, place recognition, and robot localization in challenging aquatic environments. His work addresses one of the most pressing obstacles in underwater autonomy: enabling robots to navigate reliably without GPS, in turbid, acoustically complex conditions where traditional sensors fail. Santos has made significant contributions to the development of topological descriptors for forward-looking sonar (FLS) images, pioneering methods that allow autonomous underwater vehicles (AUVs) to recognize previously visited locations using acoustic imagery alone. His most cited work, "Underwater Place Recognition Using Forward-Looking Sonar Images" (2018, 40 citations), established a foundational topological framework for sonar-based navigation. He further advanced this field through deep learning approaches, including triplet network architectures for acoustic image retrieval and cross-domain matching between aerial and underwater sensor modalities. A recurring theme across his research is sensor fusion — combining sonar, aerial imagery, and neural networks to achieve robust localization under unmodeled noise and environmental uncertainty. His work on 3D underwater reconstruction using MSIS sonar further demonstrates his broad contributions to robotic perception. With over 120 cumulative citations, Santos represents a growing voice in underwater robotic autonomy research.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3A Topological Descriptor of Acoustic Images for Navigation and Mapping15 citations · 2015
- 4
- 5Underwater Sonar and Aerial Images Data Fusion for Robot Localization12 citations · 2019
- 6A modified topological descriptor for forward looking sonar images7 citations · 2016
- 7Cooperative Training of Triplet Networks for Cross-Domain Matching6 citations · 2020
- 8
- 9
- 10