Hanyang Zhuang

Shanghai Jiao Tong University, Guangxi University of Science and Technology

Papers

Total Citations

H-Index

About

Hanyang Zhuang is a researcher advancing the state of the art in robotic perception, with a primary focus on point cloud registration (PCR) and LiDAR-based SLAM. His work directly addresses critical bottlenecks in autonomous navigation, particularly the need for algorithms that are both highly accurate and computationally efficient. Zhuang’s major contributions include the development of **G3DOA**, a generalizable 3D descriptor that leverages overlap attention to robustly identify and match points in overlapping regions between point clouds, a fundamental challenge for PCR. This work has garnered 13 citations for its novel approach to creating descriptors that generalize across different environments. He has also pioneered the **GOGICP** method, a real-time Gaussian Octree-based GICP algorithm that achieves a crucial balance between high registration accuracy and real-time performance, a persistent conflict in classical methods. Furthermore, his research on recognizing degradation scenarios for LiDAR SLAM systems addresses a practical vulnerability, enabling robots to detect and mitigate localization failures in sparse, featureless environments. Through these contributions, Zhuang is building more reliable and efficient perception systems for autonomous driving and robotics.