Zehui Yuan

Politecnico di Torino, North University of China

Papers

Total Citations

H-Index

About

Zehui Yuan is a researcher whose work bridges the critical intersection of robotic perception and efficient computational optimization. His primary contributions lie in two key areas: advancing graph-based optimization for robotics and developing robust methods for human-robot interaction. In his influential 2012 paper, "Graph Optimization with Unstructured Covariance," Yuan introduced a fast, accurate linear approximation technique that significantly improved the efficiency of solving complex optimization problems, a foundational contribution for modern SLAM systems. This work, with 6 citations, remains a reference point for researchers seeking computational speed without sacrificing precision. Yuan also tackled the challenging problem of people detection and tracking from small-footprint ground robots. His 2018 paper presents an innovative lower part-based approach, enabling reliable perception from a low-lying viewpoint where traditional methods fail due to limited visible features. This work, with 4 citations, addresses a practical bottleneck in deploying robots in crowded indoor environments. By combining theoretical rigor with applied solutions for real-world robotic challenges, Yuan’s research has directly enhanced both the accuracy and practicality of autonomous systems, making him a notable figure in the field of field robotics and perception.