Max Robotics — Global Service Robot Directory + US Certification

About

Guanqi Zhan is a rising researcher in 3D computer vision and robotics, whose work focuses on enabling intelligent agents to perceive, assemble, and manipulate objects in complex, real-world environments. His key research areas include 3D part assembly, articulated object manipulation, and language-guided navigation. Zhan’s most impactful contribution is his pioneering work on **Generative 3D Part Assembly via Dynamic Graph Learning** (2020, 46 citations), which tackled the challenging task of autonomously assembling 3D parts—analogous to assembling IKEA furniture—by modeling part relationships as a dynamic graph. He further advanced this field with **Score-PA** (2023), introducing a generative, score-based approach that eliminates the need for predefined assembly instructions. In manipulation, Zhan’s **Learning Environment-Aware Affordance** (2023) addresses the critical problem of occlusions in articulated object manipulation, providing actionable priors for home-assistant robots. Most recently, his work **InstructNav** (2024) enables zero-shot navigation from generic language instructions in unexplored environments, a significant step toward practical human-robot interaction. With a growing citation count and a clear trajectory from foundational assembly to real-world navigation, Zhan is establishing himself as a key innovator in embodied AI and autonomous robotics.