Kai Luan

National University of Defense Technology

Papers

Total Citations

H-Index

About

Kai Luan is a researcher specializing in all-weather perception for autonomous systems, with a primary focus on millimeter-wave (mmWave) radar sensing and point cloud enhancement. His most notable contribution is the development of diffusion-based super-resolution techniques that address the critical limitations of mmWave radar data—namely, its inherent sparsity and contamination by ghost points. In his 2024 paper, "Diffusion-Based Point Cloud Super-Resolution for mmWave Radar Data," Luan introduced a novel generative framework that significantly improves radar point cloud density and accuracy, enabling more reliable perception in adverse weather conditions where cameras and LiDAR fail. This work, already garnering 14 citations, demonstrates his ability to bridge the gap between cutting-edge generative AI and practical robotics challenges. Luan’s research is particularly impactful for outdoor mobile robotics, autonomous driving, and any application requiring robust sensing in fog, rain, or snow. By tackling the fundamental data quality issues of mmWave radar, he is paving the way for safer, more resilient autonomous systems. His work stands out for its innovative use of diffusion models—a state-of-the-art generative approach—to solve a real-world engineering problem, marking him as a rising figure in the field of perception for robotics.