Lan Du

PLA Information Engineering University, Monash University

Papers

Total Citations

H-Index

About

Lan Du is a researcher whose work bridges the frontiers of astronomical instrumentation and efficient deep learning. In astronomical surveying, Du pioneered an automatic method using video measurement robots and CCD imaging technology. This innovation eliminates the subjective errors of traditional celestial geodesy—the so-called personal and instrumental equation—enabling high-precision, fully automated measurements. This foundational work has garnered 9 citations, marking a significant step toward modernizing astrometric observation. Simultaneously, Du contributes to the rapidly evolving field of model compression. By developing a feature fusion-based collaborative learning approach for knowledge distillation, Du addresses a critical challenge: training highly efficient deep neural networks without sacrificing performance. This technique is vital for deploying advanced AI in resource-constrained environments like self-driving cars and intelligent robotics. Though a newer contribution with 2 citations, it reflects a forward-looking engagement with the practical deployment of AI. Du’s dual focus—enhancing both our view of the cosmos and the efficiency of the algorithms that interpret it—demonstrates a versatile and impactful research portfolio, tackling fundamental measurement challenges in both physical and digital domains.