Chien Erh Lin
Papers
2
Total Citations
25
H-Index
2
About
Chien Erh Lin is a rising star in robotics whose work sits at the intersection of geometry, symmetry, and learning for perception and control. His research focuses on exploiting the inherent symmetry structures of physical systems—particularly the SE(3) group of rigid motions—to build more robust, data-efficient, and generalizable robot algorithms. In his landmark survey, "Progress in symmetry preserving robot perception and control through geometry and learning" (2022, 13 citations), Lin synthesized how geometric tools can unify sensor registration, state estimation, and control by respecting the symmetries of the problem. His most impactful contribution to date is the **SE3ET: SE(3)-Equivariant Transformer** (2024, 12 citations), a novel architecture designed for low-overlap point cloud registration. By enforcing equivariance to rotations and translations directly in the network’s layers, SE3ET dramatically improves registration accuracy under large initial pose errors—a notoriously difficult scenario in robotics. This work has quickly become a reference for researchers tackling partial point cloud alignment. Lin’s achievements demonstrate a rare ability to bridge deep learning with classical geometric principles, offering a principled path toward more reliable and interpretable robot perception systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2