Minghu Wu

Papers

Total Citations

H-Index

About

Minghu Wu is a researcher at the forefront of intelligent systems for energy infrastructure and wearable robotics. His work primarily focuses on applying advanced deep learning and computer vision techniques to solve critical challenges in overhead transmission line maintenance and lower limb exoskeleton control. Wu’s most cited paper, “OTL-Classifier: Towards Imaging Processing for Future Unmanned Overhead Transmission Line Maintenance” (2019, 16 citations), introduces a pioneering approach to automated inspection of power lines, addressing the growing global demand for reliable electricity by enabling unmanned aerial vehicle-based diagnostics. This work has laid important groundwork for safer, more efficient maintenance of critical energy infrastructure. More recently, Wu has advanced human-robot interaction with his 2024 paper “A SE-DenseNet-LSTM model for locomotion mode recognition in lower limb exoskeleton” (7 citations), which proposes a novel hybrid architecture combining DenseNet, LSTM, and channel attention mechanisms. This model significantly improves the accuracy of recognizing human locomotion modes—such as walking, stair climbing, and running—enabling more intuitive and responsive control of wearable exoskeletons. By bridging computer vision, time-series analysis, and attention mechanisms, Wu’s research demonstrates a clear trajectory from infrastructure monitoring to assistive robotics, with tangible impact on both industrial maintenance and rehabilitation technology.