Guangyun Xu

Chinese Academy of Sciences, Shandong Institute of Automation, University of Chinese Academy of Sciences

Papers

Total Citations

H-Index

About

Guangyun Xu is a leading researcher in robotic dexterous manipulation, with a primary focus on enabling robots to perform human-like grasping and picking in complex, cluttered environments. His work bridges computer vision and robotics, particularly through the innovative use of point cloud data and deep learning. Xu’s most influential paper, “Robotics Dexterous Grasping: The Methods Based on Point Cloud and Deep Learning” (2021, 58 citations), provides a foundational survey that has shaped the field’s understanding of how robots can achieve precise, adaptive grasps. He further advanced the state of the art with “GPR: Grasp Pose Refinement Network for Cluttered Scenes” (2021, 32 citations), introducing a network that refines grasp poses by incorporating local geometry awareness—a critical improvement over single-shot detection methods. His work on “POIS: Policy-Oriented Instance Segmentation for Ambidextrous Robot Picking” (2021) demonstrates a novel approach to coordinating parallel-jaw grippers and suction cups, enabling efficient, policy-driven picking. With over 90 combined citations, Xu’s contributions are pivotal for students and researchers aiming to develop more capable, human-assistive robotic systems for industrial and daily-life applications.