top of page

>Return

Sensor fusion & data analytics/machine learning for smart systems

(1) Data Analytics and Machine Learning:

SMART Lab applies advanced data analytics techniques for the quality assurance of smart additive manufacturing, including anomaly detection and hyperparameter tuning using multi-task Gaussian processes. The research includes developing novel attention-aware deep learning models for various applications, such as COVID-19 diagnosis from CT scan images and defect detection in manufacturing processes.

Representative Papers:

  1. Chung J., Kong Z.J. "Imbalanced Data Classification via Generative Adversarial Network with Application to Anomaly Detection in Additive Manufacturing Process." Journal of Intelligent Manufacturing, 2023. DOI: 10.1007/s10845-023-02163-8.

  2. Shen B., Xie W., and Kong Z.J. "Smooth Robust Tensor Completion for Background/Foreground Separation with Missing Pixels: Novel Algorithm with Convergence Guarantee." Journal of Machine Learning Research, Vol. 23, No. 217, pp. 1-40, 2022.

  3. Gnanasambandam, R.*, Shen, B.*, Chung, J.*, Yue, X., Kong, Z.J., 2023. Self-scalable Tanh (Stan): Multi-Scale Solutions for Physics-Informed Neural Networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 45, No. 12, pp. 15588-15603. DOI: 10.1109/TPAMI.2023.3307688.

  4. Shen B., Kamath R.R., Choo H., and Kong Z.J. "Robust tensor decomposition based background/foreground separation in noisy videos and its applications in additive manufacturing." IEEE Transactions on Automation Science and Engineering, Vol. 20, No. 1, pp. 583-596, 2022. DOI: 10.1109/TASE.2022.3163674.

  5. Maftouni, M.*, Shen, B.*, Law, A.*, Yazdib, N., Hadavandc, F., Ghiasvandb, F, and Kong, Z.J., 2022. A Mask-guided Attention Deep Learning Model for COVID-19 Diagnosis based on an Integrated CT Scan Images Database. IISE Transactions on Healthcare Systems Engineering, DOI: 10.1080/24725579.2022.2142866.

(2) Additive Manufacturing (AM):

SMART Lab's work in additive manufacturing focuses on developing online real-time quality monitoring systems using heterogeneous sensors to ensure the integrity and quality of additive manufacturing processes. The lab employs reinforcement learning-based defect mitigation and quality assurance frameworks and utilizes deep learning models for in-situ process monitoring and surface morphology measurement.

Representative Papers:

  1. Chung J., Shen B., Law A., and Kong Z.J. "Reinforcement Learning-based Defect Mitigation for Quality Assurance of Additive Manufacturing." Journal of Manufacturing Systems, Vol. 65, pp. 822-835, 2022. DOI: 10.1016/j.jmsy.2022.11.008.

  2. Wang R., Garcia D., Law A., and Kong Z.J. "Development of Structured Light 3D-Scanner with High Spatial Resolution and its Applications for Additive Manufacturing Quality Assurance." International Journal of Advanced Manufacturing Technology, Vol. 117, No. 3, pp. 845-862, 2021. DOI: 10.1007/s00170-021-07780-2.

  3. Liu C., Wang R., Ho I., Kong Z.J., Williams C.B., Babu S., and Joslin C. "Toward Online Layer-wise Surface Morphology Measurement in Additive Manufacturing Using a Deep Learning-based Approach." Journal of Intelligent Manufacturing, 2022. DOI: 10.1007/s10845-022-01933-0.

  4. Rao P., Liu J., Roberson D., Kong Z.J., and Williams C. "Online Real-time Quality Monitoring in Additive Manufacturing Processes using Heterogeneous Sensors." ASME Transactions Journal of Manufacturing Science and Engineering, Vol. 137, No. 6, pp. 1007-1 - 1007-12, 2015. DOI: 10.1115/1.4029823.

(3) Sensing Technologies:

SMART Lab's research in sensing technologies includes the use of multi-sensing and correlation analysis for in-situ melt pool measurements in laser powder bed fusion, a crucial process in additive manufacturing. The lab also focuses on high-resolution sub-surface thermal measurement using novel fiber optics combined with machine learning for better process control and optimization.

Representative Papers:

  1. Wang R., Wang R., Dou C., Yang S., Gnanasambandam R., Wang A., and Kong Z.J. "Sub-surface thermal measurement in additive manufacturing via machine learning-enabled high-resolution fiber optic sensing." Nature Communications, 2024 (accepted).

  2. Wang Y., Wang R., Yue X., and Kong Z.J. "Structured Light Scanning Based 3D Scanning for Process Monitoring and Quality Control in Precast/Prestressed Concrete Production." PCI Journal, Vol. 66, No. 6, pp. 17-32, 2021.

  3. Shen B., Wang R., Law A., Kamath R., Choo H., and Kong Z.J. "Super Resolution for Multi-Sources Image Stream Data using Smooth and Sparse Tensor Completion and its Applications in Data Acquisition of Additive Manufacturing." Technometrics, Vol. 64, No. 1, pp. 2-17, 2021. DOI: 10.1080/00401706.2021.1905074.

  4. Wang R., Garcia D., Kamath R.R., Dou C., Ma X., Shen B., Choo H., Fezzaa K., Yu H.Z., and Kong Z.J. "In Situ Melt Pool Measurements for Laser Powder Bed Fusion using Multi Sensing and Correlation Analysis." Scientific Reports, Vol. 12, No. 1, pp. 1-17, 2022. DOI: 10.1038/s41598-022-18096-w.

(4) Robust and Adaptive Systems:

SMART Lab's contributions to robust and adaptive systems include the creation of robust tensor decomposition methods for background/foreground separation in noisy video data, applicable to additive manufacturing monitoring. The lab also implements Bayesian optimization and sparse Bayesian learning techniques for fault diagnosis and quality control in manufacturing systems.

Representative Papers:

  1. Gnanasambandam R., Shen B., Law A.C.C., Dou C., and Kong Z. "Deep Gaussian Process for Enhanced Bayesian Optimization and its Application in Additive Manufacturing." IISE Transactions, 2024 (accepted). DOI: 10.1080/24725854.2024.2312905.

  2. Shen B., Gnannasambandam R., Wang R., and Kong Z.J. "Multi-Task Gaussian Process Upper Confidence Bound for Hyperparameter Tuning and its Application for Simulation Studies of Additive Manufacturing." IISE Transactions, 2022.

  3. Li Y., Shi Z., Liu C., Tian W., Kong Z.J., and Williams C.B. "Augmented Time Regularized Generative Adversarial Network (ATR-GAN) for Data Augmentation in Online Process Anomaly Detection." IEEE Transactions on Automation Science and Engineering, Vol. 19, No. 4, pp. 3338-3355, 2021. DOI: 10.1109/TASE.2021.3118635.

  4. Shen B., Kamath R.R., Choo H., and Kong Z.J. "Robust tensor decomposition based background/foreground separation in noisy videos and its applications in additive manufacturing." IEEE Transactions on Automation Science and Engineering, Vol. 20, No. 1, pp. 583-596, 2022. DOI: 10.1109/TASE.2022.3163674.

These papers illustrate SMART Lab's significant contributions to the fields of AI, machine learning, sensing, and smart manufacturing.

bottom of page