Max Robotics — Global Service Robot Directory + US Certification

About

Juzhan Xu is a researcher working at the intersection of computer vision, robotics, and reinforcement learning, with a particular focus on 3D spatial reasoning and autonomous robot planning. His work addresses some of the most practically demanding challenges in embodied AI, including 3D bin packing, shape orientation estimation, and motion planning for robotic systems. Xu's most notable contributions include the Neural Packing framework, which presents an end-to-end solution to the transport-and-packing problem by integrating RGBD sensing, object recognition, and reinforcement learning into a unified pipeline — a significant step toward deployable robotic automation. Complementing this, his deliberate planning approach to 3D bin packing using configuration trees tackles real-world industrial constraints that earlier methods struggled to handle. His UprightRL work reframes 3D shape orientation estimation as a sequential decision-making problem, demonstrating creative cross-domain thinking. More recently, his PC-Planner introduces physics-constrained self-supervised learning for robust neural motion planning, addressing high-dimensional challenges central to modern embodied AI. With papers accumulating citations across robotics and computer vision venues, Xu's research reflects a consistent drive to bridge perception and physical reasoning — making him a rising contributor to intelligent robotic systems research.