Max Robotics — Global Service Robot Directory + US Certification

About

Xijun Zhao is a robotics and computer vision researcher whose work spans autonomous navigation, scene understanding, and robust localization for intelligent robotic systems. His research concentrates on three interconnected domains: semantic scene segmentation for off-road environments, visual simultaneous localization and mapping (SLAM), and stereo vision for depth perception. Zhao's most impactful contribution is his work on fine-grained off-road semantic segmentation using contrastive learning, which advances beyond traditional binary road classification to enable nuanced scene understanding for mobile robots navigating complex outdoor terrain — earning 30 citations since 2021. He has also made persistent contributions to visual SLAM in weakly textured environments through his RWT-SLAM system, a robust framework that addresses one of the field's most persistent challenges, accumulating citations across multiple publications. His 2022 work introducing a normalized disparity loss for stereo matching networks reflects a thoughtful approach to improving deep learning training for robotic vision applications. Earlier research on scene-adaptive LiDAR covariance error modeling demonstrates his grounding in sensor fusion and probabilistic localization. Collectively, Zhao's portfolio reflects a consistent commitment to making robotic perception more reliable across challenging real-world conditions.