Guilherme N. DeSouza
Purdue University West Lafayette, University of Missouri, The University of Western Australia
Papers
16
Total Citations
1,748
H-Index
9
About
Guilherme N. DeSouza is a robotics and computer vision researcher whose work spans mobile robot navigation, agricultural robotics, and intelligent control systems. He is perhaps best known for his landmark 2002 survey, "Vision for Mobile Robot Navigation," which synthesized two decades of progress in the field and has since accumulated over 1,360 citations, cementing its status as a foundational reference for researchers worldwide. His contributions extend beyond navigation theory into practical robotic systems: his development of the Vinobot and Vinoculer platforms introduced an innovative dual-robot architecture for high-throughput field phenotyping, addressing critical challenges in precision agriculture and earning over 140 citations since 2017. DeSouza has also made meaningful advances in robot kinematics, proposing neural network and fuzzy logic approaches to solving inverse kinematics problems, as well as parallel-processing numerical methods for general serial manipulators. His broader research portfolio includes vision-based robotic architectures, stereo head calibration, 3D object modeling, airborne target geolocation, and embedded hardware solutions for assistive technology. Across these diverse areas, DeSouza consistently bridges theoretical rigor with real-world application, making his work highly relevant to both academic researchers and practitioners building the next generation of intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1Vision for mobile robot navigation: a survey1,360 citations · 2002
- 2
- 3
- 4
- 5
- 63-D Modeling of Real-World Objects Using Range and Intensity Images23 citations · 2005
- 7
- 8Geolocation of Multiple Targets from Airborne Video Without Terrain Data19 citations · 2010
- 9
- 10