Marco Caravagna

Papers

Total Citations

H-Index

About

Marco Caravagna is a leading researcher in robotics and state estimation, with a primary focus on enhancing the localization and perception capabilities of humanoid robots. His work bridges the gap between proprioceptive and exteroceptive sensing, addressing the critical challenge of drift accumulation in robotic navigation. Caravagna’s most notable contribution is his development of an overlap-based Iterative Closest Point (ICP) tuning method, which significantly improves the robustness of laser-based scene registration for humanoid platforms. This approach, detailed in his highly cited 2017 paper (31 citations), corrects positional drift by intelligently filtering out non-overlapping point cloud regions, thereby enabling more reliable localization in dynamic environments. His research has direct implications for real-world applications, from disaster response to autonomous exploration, where precise robot positioning is essential. Caravagna’s work is widely recognized for its practical impact, offering a scalable solution to a fundamental problem in mobile robotics. By advancing state estimation techniques, he continues to shape the future of humanoid robot autonomy, making his contributions a cornerstone for students and researchers in the field.