Max Robotics — Global Service Robot Directory + US Certification

About

Yandong Guo is a computer vision and robotics researcher whose work bridges visual localization, scene understanding, and intelligent robotic manipulation. His research spans two interconnected frontiers: high-precision camera re-localization and vision-language-action models for robotic systems. In the domain of visual localization, Guo has made notable contributions by addressing the limitations of point-only re-localization approaches. His 2022 work on joint optimization of visual points and lines demonstrated meaningful advances in camera pose refinement accuracy, garnering 18 citations and establishing itself as his most recognized contribution. His integrated re-localization framework RLOCS (2021) further combined image retrieval with semantic constraints to tackle challenging real-world conditions, while his real-time fusion framework extended these capabilities to long-term, degraded environments. More recently, Guo has turned his attention to robotic reasoning and manipulation. His RoboMamba model introduces efficient Vision-Language-Action architectures leveraging state-space models to improve both reasoning capacity and execution speed in robotic agents. His 2025 Fast-in-Slow framework further explores dual-system designs that unify rapid manipulation with deliberate reasoning, reflecting a growing ambition to build generalist robotic policies. Across his career, Guo's work consistently targets real-world deployment challenges in augmented reality, autonomous driving, and robotics.