Papers
5
Total Citations
68
H-Index
4
About
Driven by the convergence of artificial intelligence and advanced manufacturing, Junliang Wang’s research focuses on 3D point cloud processing, robotic manipulation, and intelligent defect detection. His most influential work introduces an unequal deep learning approach for 3D point cloud segmentation, challenging the conventional assumption that all points are equally important—a critical insight for autonomous driving and robotic navigation. This paper has garnered 38 citations and established a new paradigm for boundary-aware segmentation. Wang further extends this expertise to practical manufacturing, developing a welding path planning method that integrates point cloud data with deep learning to automate robotic welding, earning 20 citations. In smart manufacturing, he addresses the challenge of surface defect detection with limited samples through a dual-metric neural network with attention guidance, a human-centric innovation for quality control. His recent work on physics-informed graph learning for shape prediction of deformable linear objects—such as cables and wires—pushes boundaries in robotics and aerospace. With contributions spanning from foundational segmentation algorithms to applied manufacturing systems, Wang’s research demonstrates a clear trajectory from theoretical innovation to real-world impact, making him a notable figure in intelligent robotics and manufacturing.
Research Focus
Key Achievements
Top Papers
- 1An Unequal Deep Learning Approach for 3-D Point Cloud Segmentation38 citations · 2020
- 2A new welding path planning method based on point cloud and deep learning20 citations · 2020
- 3
- 4Shape reconstruction based on FBG flexible sensor4 citations · 2022
- 5