Lucas Nunes

University of Bonn, Universidade de São Paulo

Papers

Total Citations

H-Index

About

Lucas Nunes is a leading researcher in robotics, specializing in 3D mapping, autonomous navigation, and place recognition for dynamic and unstructured environments. His work addresses critical challenges in enabling robots to operate reliably in real-world settings, from urban streets to agricultural fields. Nunes is best known for developing ERASOR2, an instance-aware mapping framework that robustly separates static environments from dynamic objects, achieving 34 citations and setting a new standard for long-term robot localization. He also introduced KPPR, a momentum contrast-based method for point cloud place recognition (16 citations), which significantly improves loop closure detection in SLAM systems. Earlier in his career, Nunes contributed foundational work on path planning and autonomous navigation using AMCL and AD, with 19 citations. Most recently, his pioneering research on zero-shot semantic segmentation for agricultural robots (2025) tackles the pressing need for sustainable weed control, enabling robots to generalize across unseen crop types without retraining. With a growing citation impact and a clear trajectory toward practical, deployable solutions, Nunes is shaping the future of robust, intelligent robotic perception.