Gaofei Xu

Shenyang Institute of Automation, Chinese Academy of Sciences

Papers

Total Citations

H-Index

About

Gaofei Xu is a researcher at the forefront of underwater robotics, specializing in intelligent perception, health monitoring, and autonomous navigation for marine environments. His work bridges the gap between advanced machine learning and the harsh realities of subsea operations. Xu’s most cited paper, “Online Learning Based Underwater Robotic Thruster Fault Detection” (2021, 13 citations), introduces a novel, real-time algorithm that models the relationship between control variables and propeller speed, enabling early detection of thruster failures—a critical contribution to the safety and longevity of autonomous underwater vehicles (AUVs). Building on this, he has pioneered lightweight computer vision architectures, such as a temporal-feature-enhanced YOLO network for high-resolution sonar segmentation (2025, 5 citations), designed to overcome the computational constraints of AUVs in dynamic environments. His work also extends to deep-sea archaeology, where a deep aggregation network with deformation convolution (2023, 2 citations) enables robust detection of marine cultural artifacts under poor imaging conditions. By integrating online learning, efficient neural networks, and domain-specific feature engineering, Xu is advancing the reliability and perceptual intelligence of next-generation underwater robotic systems.