Max Robotics — Global Service Robot Directory + US Certification

About

Kuan Xu is a robotics researcher whose work bridges the gap between deep learning and physics-based optimization for robust robot perception and navigation. His primary research areas include visual simultaneous localization and mapping (SLAM), visual odometry, and robot learning. Xu’s most impactful contribution is **AirSLAM** (2025, 64 citations), an efficient point-line visual SLAM system that tackles both short- and long-term illumination challenges by hybridizing deep learning feature detection with traditional methods. He is also the lead developer of **PyPose** (2023, 35 citations), a widely adopted library that integrates physics-based optimization with deep learning, enabling robots to generalize better in changing environments. Additionally, Xu introduced a fast, non-iterative RGB-D visual odometry method (2024) that leverages planar elements for efficient 6-DoF pose estimation. His work is notable for its practical focus on real-time performance and robustness, making it highly relevant for autonomous systems operating in challenging conditions. With a growing citation record, Xu is establishing himself as a key figure in advancing SLAM and robot learning.