Xianfei Pan

National University of Defense Technology

Papers

Total Citations

103

H-Index

About

Xianfei Pan is a leading researcher in the fields of inertial navigation, deep learning for positioning, and cooperative robotics. His most impactful work, the 2024 survey "Deep Learning for Inertial Positioning," has already garnered 95 citations, establishing him as a key voice in the integration of neural networks with ubiquitous sensor-based localization for smartphones, drones, and vehicles. Pan’s major contributions include advancing self-supervised learning for autonomous systems, as demonstrated in "SelfOdom," where he pioneered a bi-directional coarse-to-fine scale recovery method for egomotion and depth estimation from monocular images—a critical step toward reducing reliance on expensive labeled data. Earlier in his career, he tackled the challenge of multi-robot coordination with "Fully Decentralized Cooperative Localization," proposing an efficient, centralized-equivalent solution for robot teams. This work highlights his sustained focus on scalable, real-world localization systems. With a research portfolio that bridges theoretical innovation and practical deployment, Pan’s work is essential reading for students and engineers working on autonomous navigation, sensor fusion, and deep learning for robotics.