Max Robotics — Global Service Robot Directory + US Certification

About

Rodrigo Marcuzzi is a leading researcher in 3D perception for autonomous systems, with a focus on large-scale mapping, LiDAR processing, and agricultural robotics. His work tackles the critical challenge of enabling robots to understand and navigate complex outdoor environments using sparse, real-world sensor data. Marcuzzi’s most influential contribution is the "Retriever" framework (26 citations), which pioneered efficient place recognition directly on compressed 3D maps—a breakthrough for autonomous driving and long-term robotic deployments where memory and bandwidth are limited. He further advanced the field with "Make it Dense" (20 citations), a self-supervised method that completes sparse LiDAR scans into dense geometric models, dramatically improving the reliability of outdoor mapping systems. Demonstrating the versatility of his approach, Marcuzzi also developed a transformer-based architecture for 3D fruit shape completion (13 citations), enabling agricultural robots to accurately estimate occluded fruit volumes from partial LiDAR views. By bridging compression, completion, and reconstruction, Marcuzzi’s work has set new standards for efficient, robust 3D perception, earning him recognition as a key innovator in both autonomous navigation and precision agriculture.