Embedded Files
Physics-Informed Machine Learning for Industrial & Physical Systems
Physics-Informed Machine Learning for Industrial & Physical Systems
Dr. Iljoo Jeong is a Senior Researcher and Team Lead of the AI Application Research Team at the Korea Electronics Technology Institute (KETI). He works on AI for sensing and diagnostics in industrial systems, using acoustic, ultrasonic, radar, and vision data to detect faults, predict quality, and support decisions in production. He received his Ph.D. in Mechanical Engineering from POSTECH, where his doctoral work included collaborative research with KRISS and KIMM, and completed an intensive AI program at Carnegie Mellon University on a government scholarship. He previously worked in industry at ASML and Mando.
Research Interests: physics-informed machine learning, inverse problems, acoustic & multimodal sensing, manufacturing diagnostics & PHM, and physical AI
Research
Research
My research develops physics-informed machine learning methods for sensing, diagnosis, and prediction in industrial and physical systems.
Acoustic & Multimodal Sensing
Using acoustic, vibration, radar, and multimodal measurements to infer sources, environments, and human activities.
Sound-source localization · Acoustic room-geometry estimation · Radar-based human sensing · Multimodal inverse problems
Manufacturing Diagnostics & PHM
Developing AI methods for fault diagnosis, quality prediction, and equipment health monitoring in manufacturing systems.
Wafer-map failure-pattern analysis · Tool-wear assessment · Machine-condition diagnosis · Product-quality prediction
Toward Intelligent Manufacturing Systems
Extending sensing, diagnosis, and prediction toward more reliable and autonomous manufacturing systems through physics-informed modeling and industrial AI.
Selected Publications
Selected Publications
EchoScan: Scanning Complex Room Geometries via Acoustic Echoes — IEEE/ACM TASLP 2024
Dual-port Conditional Invertible Neural Network for Sound Intensity Compensation in Sound Source Localization — IEEE TIM 2024
Wafer Map Failure Pattern Classification using Geometric Transformation-Invariant CNN — Scientific Reports 2023
ABC-HF: Physics-grounded Human Fall Detection with mmWave — Engineering Applications of Artificial Intelligence 2025
See the Publications page for the full list of 12 journals and 20 conference papers.
Collaboration
Collaboration
Open to collaboration on joint research, technical advisory, and national R&D program planning. Representative areas:
Manufacturing data — quality and defect root-cause analysis
Sensor, acoustic, and vibration data — equipment condition diagnosis (PHM)
Physics-based modeling with AI — prediction, calibration, and optimization
Industry–academia collaboration — joint programs, on-site demonstration, and technology transfer
Contact
Contact
Email: iljjeong@keti.re.kr
Korea Electronics Technology Institute (KETI) · Manufacturing AI Division
For publications, projects, and full record, see the Publications, Projects, and Experience pages above.
Page updated
Google Sites
Report abuse