Mohammad Reza Ahmadzadeh
Papers
2
Total Citations
14
H-Index
2
About
Mohammad Reza Ahmadzadeh is a roboticist advancing autonomous navigation and perception in complex, dynamic environments. His research centers on visual simultaneous localization and mapping (SLAM), qualitative vision-based navigation, and semantic scene understanding for mobile robots. Ahmadzadeh’s most cited work, “Det-SLAM: A semantic visual SLAM for highly dynamic scenes using Detectron2” (2022, 8 citations), tackles a critical limitation of traditional SLAM systems—their fragility in environments with moving objects. By integrating Detectron2’s semantic segmentation, his approach enables robots to filter out dynamic entities, dramatically improving localization robustness. This contribution is particularly impactful for autonomous systems operating in human-populated spaces, such as service robots or autonomous vehicles. His earlier work, “Qualitative vision-based navigation based on sloped funnel lane concept” (2019, 6 citations), offers a novel, computationally efficient method for visual homing, using a “sloped funnel lane” to guide robots without precise metric maps. Together, these papers demonstrate Ahmadzadeh’s ability to blend deep learning with classical robotics principles, producing practical solutions for real-world deployment. His research is essential reading for students and engineers seeking to build robots that see, understand, and move reliably through our unpredictable world.
Research Focus
Key Achievements
Top Papers
- 1
- 2Qualitative vision-based navigation based on sloped funnel lane concept6 citations · 2019