Max Robotics — Global Service Robot Directory + US Certification

About

Marcelo Gattass is a leading researcher in mobile robotics, with a primary focus on Simultaneous Localization and Mapping (SLAM) in complex, real-world settings. His work directly tackles the critical limitation of traditional Visual SLAM algorithms, which assume static environments and fail in spaces populated by moving people or objects. Gattass’s major contribution is the development of robust SLAM systems that integrate deep learning-based object detection—specifically using models like YOLO and Mask R-CNN—to identify and filter out dynamic content. His most cited paper, "Crowd-SLAM: Visual SLAM Towards Crowded Environments using Object Detection" (2021, 78 citations), introduces a novel framework for navigating crowded spaces. This work, along with his 2019 study on the accuracy-speed trade-off in dynamic SLAM (31 citations), has been instrumental in advancing the field. By addressing the open problem of visual localization in both dynamic and changing environments, Gattass’s research is paving the way for the real-world deployment of fully autonomous mobile robots, making his contributions highly influential for students and engineers working on next-generation robotic perception.