Max Robotics — Global Service Robot Directory + US Certification

About

Ruoyu Geng is a robotics researcher whose work centers on advancing perception, mapping, and navigation for autonomous systems operating in complex, real-world environments. His primary contributions lie in multi-sensor fusion, dynamic scene understanding, and terrain reconstruction. Geng is the lead author of the highly influential **FusionPortable** benchmark (50 citations), a multi-sensor campus-scene dataset that has become a key resource for evaluating localization and mapping accuracy across diverse robotic platforms. He also pioneered a benchmark for **dynamic points removal** in point cloud maps (32 citations), addressing a critical challenge for robust localization and path planning. His work on **real-time neural dense elevation mapping** (18 citations) introduces uncertainty estimations for urban terrain, directly improving legged robot locomotion. Most recently, Geng has advanced **real-time metric-semantic mapping** (17 citations) for outdoor navigation, fusing multi-modal data to create high-level, human-prior-aware environment representations. Through these contributions, Geng has established himself as a leading voice in creating robust, perceptually aware robotic systems capable of operating safely and efficiently in dynamic, unstructured spaces.