Heonkook Kim
Papers
3
Total Citations
39
H-Index
3
About
Heonkook Kim is a leading researcher in the field of industrial robotics and intelligent fault diagnosis, with a primary focus on the early detection of cable faults in robotic systems. His work is instrumental in advancing factory automation by addressing the critical challenge of cable degradation, which is a leading cause of system downtime. Kim’s major contributions lie in developing deep learning-based diagnostic methods that can detect "soft faults"—transient, early-stage cable issues that are notoriously difficult to identify. He pioneered the use of attention recurrent neural networks and stacked transformer encoder layers to automatically extract features from robot dynamics and control signals, enabling severity estimation and fault classification without requiring balanced fault datasets. His most cited paper, "Current Only-Based Fault Diagnosis Method for Industrial Robot Control Cables" (2022), has garnered 18 citations and highlights his innovative use of novelty detection for imbalanced data scenarios. With a total of 39 citations across his top three works, Kim’s research is shaping the future of predictive maintenance, offering practical, data-driven solutions to maximize productivity in automated manufacturing environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3