Max Robotics — Global Service Robot Directory + US Certification

About

Michael Krawez is a robotics researcher whose work sits at the intersection of computer vision, semantic understanding, and robot manipulation. His primary research areas include visual place recognition, semantic grasping, and environmental illumination modeling for localization. Krawez’s most impactful contribution is **FM-Loc** (16 citations), which leverages foundation models to improve vision-based localization in changing environments—a critical challenge for SLAM systems. He further extends semantic reasoning to manipulation with **Lan-grasp** (5 citations), using large language models to enable robots to identify appropriate grasp points and natural object orientations. Krawez has also advanced real-time environmental modeling, developing methods for **outdoor illumination estimation** to stabilize indoor camera tracking and building **dense reflectance maps** of indoor scenes using RGB-D cameras. His work bridges high-level semantic understanding with low-level geometric and photometric modeling, making robots more robust to appearance variations and capable of context-aware interaction. With a growing citation record and a focus on integrating foundation models into classical robotics pipelines, Krawez is shaping the next generation of perceptive, semantically-aware autonomous systems.