Zepei Wu

National Institute of Advanced Industrial Science and Technology

Papers

Total Citations

H-Index

About

Zepei Wu is a researcher whose work sits at the intersection of robotics, computer vision, and intelligent manufacturing, with a particular focus on enabling robots to learn complex manipulation tasks. His most influential contribution is the development of a multi-modal 3D vision system that allows robots to learn object assembly directly from human demonstration. In his seminal 2017 paper, which has garnered 57 citations, Wu introduced a two-phase framework: first, a human teaches the assembly structure to the robot, and then the robot autonomously identifies, grasps, and assembles the objects. This work addresses a critical challenge in next-generation manufacturing—moving beyond rigid, pre-programmed automation toward flexible, human-guided robotic assembly. By integrating depth sensing and visual data, Wu’s approach enables robots to understand spatial relationships and perform precise manipulations in unstructured environments. His research has significant implications for industries seeking to deploy adaptable robotic systems, and his work on teaching robots through demonstration continues to influence the fields of imitation learning and human-robot collaboration.