Adaptive Hybrid Anomaly Detection for High-Duty-Cycle Industrial Robots
DOI:
https://doi.org/10.5281/zenodo.22139882Keywords:
Industrial Robot, Anomaly Detection, Condition Monitoring, Hybrid Learning, Random Forest, Principal Component Analysis, Adaptation, Predictive MaintenanceAbstract
Machine-tending applications of industrial robots remain effective after decades of use; meanwhile, controller generation becomes obsolete and difficult to manage. This study presents a controller-independent hybrid anomaly-detection algorithm for condition monitoring of a six-axis degree-of-freedom robot using an external inertial measurement unit. The online decision path combines a supervised Random Forest, which recognizes reviewed disturbance signatures, with a deterministic PCA-based novelty score referenced to context-compatible nominal sections. A bounded context correction, decision-level fusion, and an H-of-K persistence rule reduce false alarms caused by payload, speed, and transient motion changes. Offline inspection of nonlinear neighborhoods is excluded from the deployed decision rule. The harmonized archive contains 69,120 multivariate observations and 4,422 windows partitioned by physical acquisition run to prevent temporal leakage. In the controlled evaluation, the PCA branch, Random Forest, fixed-fusion model, and adaptive hybrid achieved accuracies of 91.2%, 93.5%, 95.8%, and 97.1%, with F1-scores of 0.89, 0.91, 0.94, and 0.96, respectively. These experimental results indicate that complementary supervised and geometric evidence improves discrimination while retaining traceability. The framework is lightweight and traceable and is intended for retrofit condition monitoring; it does not estimate remaining useful life. Establishing prognostic capability would require run-to-failure or overhaul data, uncertainty-aware degradation modeling, and multi-robot validation.
References
Ayankoso, S., Gu, F., Louadah, H., Fahham, H., & Ball, A. (2024). Artificial-intelligence-based condition monitoring of industrial collaborative robots: detecting anomalies and adapting to trajectory changes. Machines, 12(9), 630. https://doi.org/10.3390/machines12090630
Azzalini, D., Bonali, L., & Amigoni, F. (2021). A minimally supervised approach based on variational autoencoders for anomaly detection in autonomous robots. IEEE Robotics and Automation Letters, 6(2), 2985-2992. https://doi.org/10.1109/LRA.2021.3062597
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5-32. https://doi.org/10.1023/a:1010933404324
Chirayil Nandakumar, S., Mitchell, D., Erden, M. S., Flynn, D., & Lim, T. (2024). Anomaly detection methods in autonomous robotic missions. Sensors, 24(4), 1330. https://doi.org/10.3390/s24041330
Choi, K., Yi, J., Park, C., & Yoon, S. (2021). Deep learning for anomaly detection in time-series data: Review, analysis, and guidelines. IEEE Access, 9, 120043-120065. https://doi.org/10.1109/ACCESS.2021.3107975
Choi, S.-H., An, D., Lee, I., & Lee, S. (2024). Anomaly detection based on graph convolutional network–variational autoencoder model using time-series vibration and current data. Mathematics, 12(23), 3750. https://doi.org/10.3390/math12233750
Deng, A., & Hooi, B. (2021). Graph neural network-based anomaly detection in multivariate time series. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 35, pp. 4027-4035). https://doi.org/10.1609/aaai.v35i5.16523
Fedorova, D., Tlach, V., Kuric, I., Dodok, T., Zajačko, I., & Tucki, K. (2024). Technical diagnostics of industrial robots using vibration signals: Case study on detecting base unfastening. Applied Sciences, 15(1), 270. https://doi.org/10.3390/app15010270
Fordal, J. M., Schjølberg, P., Helgetun, H., Skjermo, T. Ø., Wang, Y., & Wang, C. (2023). Application of sensor data based predictive maintenance and artificial neural networks to enable Industry 4.0. Advances in Manufacturing, 11(2), 248-263. https://doi.org/10.1007/s40436-022-00433-x
Garg, A., Zhang, W., Samaran, J., Savitha, R., & Foo, C.-S. (2022). An evaluation of anomaly detection and diagnosis in multivariate time series. IEEE Transactions on Neural Networks and Learning Systems, 33(6), 2508-2517. https://doi.org/10.1109/tnnls.2021.3105827
Huang, Y., Yang, B., Wang, S., Xie, K., Wang, Y., Yi, L., & Dong, N. (2025). Attentive neural processes based on reliable inferences for industrial equipment anomaly detection. Computers in Industry, 171, 104331. https://doi.org/10.1016/j.compind.2025.104331
ISO. (2003). 13374-1:2003—Condition Monitoring and Diagnostics of Machines: Data Processing, Communication and Presentation—Part 1: General Guidelines. In (1 ed.). International Organization for Standardization. https://www.iso.org/standard/21832.html
ISO. (2018). ISO 17359:2018—Condition Monitoring and Diagnostics of Machines: General Guidelines. In. International Organization for Standardization. https://www.iso.org/standard/71194.html
Jolliffe, I. T., & Cadima, J. (2016). Principal component analysis: a review and recent developments. Philosophical transactions. Series A, Mathematical, physical, and engineering sciences, 374(2065), 20150202. https://doi.org/10.1098/rsta.2015.0202
Kayan, H., Heartfield, R., Rana, O., Burnap, P., & Perera, C. (2024). Casper: Context-aware iot anomaly detection system for industrial robotic arms. ACM Transactions on Internet of Things, 5(3), 1-36. https://doi.org/10.1145/3670414
Kermenov, R., Nabissi, G., Longhi, S., & Bonci, A. (2023). Anomaly detection and concept drift adaptation for dynamic systems: a general method with practical implementation using an industrial collaborative robot. Sensors, 23(6), 3260. https://doi.org/10.3390/s23063260
Khan, Z., Nasir, A., & Mekid, S. (2025). Fault-tolerant control strategies for industrial robots: State of the art and future perspective on AI-based fault management. Artificial Intelligence Review, 58(11), 362. https://doi.org/10.1007/s10462-025-11327-2
Kim, H. (2025). Fault Diagnosis in Robot Drive Systems Using Data-Driven Dynamics Learning. In Actuators (Vol. 14, pp. 583). MDPI. https://doi.org/10.3390/act14120583
Kim, S., Choi, K., Choi, H.-S., Lee, B., & Yoon, S. (2022). Towards a rigorous evaluation of time-series anomaly detection. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 36, pp. 7194-7201). https://doi.org/10.1609/aaai.v36i7.20680
Kumar, P., Khalid, S., & Kim, H. S. (2023). Prognostics and health management of rotating machinery of industrial robot with deep learning applications—A review. Mathematics, 11(13), 3008. https://doi.org/10.3390/math11133008
Kumar, P., Raouf, I., & Kim, H. S. (2024). Transfer learning for servomotor bearing fault detection in the industrial robot. Advances in engineering software, 194, 103672. https://doi.org/10.1016/j.advengsoft.2024.103672
Li, J., Chen, Y., & Xing, Y. (2023). Memory mechanism for unsupervised anomaly detection. In The 39th Conference on Uncertainty in Artificial Intelligence. https://openreview.net/pdf?id=8xPj4G7E-yG
Liu, Q., Yu, Y., Han, B. S., & Zhou, W. (2025). An evolutionary deep learning approach using flexible variable-length dynamic stochastic search for anomaly detection of robot joints. Applied Soft Computing, 168, 112493. https://doi.org/10.1016/j.asoc.2024.112493
Lu, K., Chen, C., Wang, T., Cheng, L., & Qin, J. (2022). Fault diagnosis of industrial robot based on dual-module attention convolutional neural network. Autonomous Intelligent Systems, 2(1), 12. https://doi.org/10.1007/s43684-022-00031-5
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In (Vol. 30). Advances in Neural Information Processing Systems. https://tinyurl.com/ye26vpcd
Mallioris, P., Aivazidou, E., & Bechtsis, D. (2024). Predictive maintenance in Industry 4.0: A systematic multi-sector mapping. CIRP journal of manufacturing science and technology, 50, 80-103. https://doi.org/10.1016/j.cirpj.2024.02.003
Niu, Q., Sui, Z., Han, J., & Zhao, Y. (2025). An industrial robot gearbox fault diagnosis approach using multi-scale empirical mode decomposition and a one-dimensional convolutional neural network-bidirectional gated recurrent unit method. Processes, 13(6), 1722. https://doi.org/10.3390/pr13061722
Pan, J., Qu, L., & Peng, K. (2022). Deep residual neural-network-based robot joint fault diagnosis method. Scientific reports, 12(1), 17158. https://doi.org/10.1038/s41598-022-22171-7
Pang, G., Shen, C., Cao, L., & Hengel, A. V. D. (2021). Deep learning for anomaly detection: A review. ACM Computing Surveys (CSUR), 54(2), 1-38. https://doi.org/10.1145/3439950
Paparrizos, J., Boniol, P., Palpanas, T., Tsay, R. S., Elmore, A. J., & Franklin, M. J. (2022). Volume Under the Surface: A New Accuracy Evaluation Measure for Time-Series Anomaly Detection. Proc. VLDB Endow., 15(11), 2774-2787. https://doi.org/10.14778/3551793.3551830
Park, J.-g., Yoon, H., & Youn, B. D. (2023). Probabilistic framework for reliable optimal design of gearboxes in general-purpose industrial robots considering random use conditions. Journal of Computational Design and Engineering, 10(2), 539-548. https://doi.org/10.1093/jcde/qwad007
Park, Y.-S., Yoo, D.-Y., & Lee, J.-W. (2021). Programmable motion-fault detection for a collaborative robot. IEEE Access, 9, 133123-133142. https://doi.org/10.1109/access.2021.3114505
Polenghi, A., Cattaneo, L., & Macchi, M. (2024). A framework for fault detection and diagnostics of articulated collaborative robots based on hybrid series modelling of Artificial Intelligence algorithms. Journal of Intelligent Manufacturing, 35(5), 1929-1947. https://doi.org/10.1007/s10845-023-02076-6
Ruff, L., Kauffmann, J. R., Vandermeulen, R. A., Montavon, G., Samek, W., Kloft, M., Dietterich, T. G., & Müller, K.-R. (2021). A unifying review of deep and shallow anomaly detection. Proceedings of the IEEE, 109(5), 756-795. https://doi.org/10.1109/JPROC.2021.3052449
Sabry, A. H., & Amirulddin, U. A. B. U. (2024). A review on fault detection and diagnosis of industrial robots and multi-axis machines. Results in Engineering, 23, 102397. https://doi.org/10.1016/j.rineng.2024.102397
Saito, T., & Rehmsmeier, M. (2015). The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS One, 10(3), e0118432. https://doi.org/10.1371/journal.pone.0118432
Schirmer, F., Kranz, P., Schmitt, J., & Kaupp, T. (2023). Anomaly detection for dynamic human-robot assembly: Application of an lstm-based autoencoder to interpret uncertain human behavior in hrc. In Companion of the 2023 ACM/IEEE International Conference on Human-Robot Interaction (pp. 333-337). https://doi.org/10.1145/3568294.3580100
Taşcı, B., Omar, A., & Ayvaz, S. (2023). Remaining useful lifetime prediction for predictive maintenance in manufacturing. Computers & Industrial Engineering, 184, 109566. https://doi.org/10.1016/j.cie.2023.109566
Tatbul, N., Lee, T. J., Zdonik, S., Alam, M., & Gottschlich, J. (2018). Precision and recall for time series. Advances in neural information processing systems, 31. https://tinyurl.com/2hpbyjjv
Tuli, S., Casale, G., & Jennings, N. R. (2022). TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data. Proceedings of the VLDB Endowment, 15(6), 1201-1214. https://doi.org/10.14778/3514061.3514067
Van der Maaten, L., & Hinton, G. (2008). Visualizing data using t-SNE. Journal of machine learning research, 9(11), 2579-2605. https://tinyurl.com/488s3pan
Vasilache, A., Nitzsche, S., Floegel, D., Schuermann, T., von Dosky, S., Bierweiler, T., Mußler, M., Kälber, F., Hohmann, S., & Becker, J. (2024). Low-Power Vibration-Based Predictive Maintenance for Industry 4.0 using Neural Networks: A Survey. arXiv:2408.00516. https://doi.org/10.48550/arXiv.2408.00516
Wang, S., Tao, J., Jiang, Q., Chen, W., Qin, C., & Liu, C. (2024). A digital twin framework for anomaly detection in industrial robot system based on multiple physics-informed hybrid convolutional autoencoder. Journal of Manufacturing Systems, 77, 798-809. https://doi.org/10.1016/j.jmsy.2024.10.016
Wu, J., Zhang, Y., Gao, B., Xia, L., Zhu, X., Wang, H., & Wan, X. (2025). A Hybrid Deep Learning Framework for Enhanced Fault Diagnosis in Industrial Robots. Algorithms, 18(12), 779. https://doi.org/10.3390/a18120779
Xu, J., Wu, H., Wang, J., & Long, M. (2022). Anomaly transformer: Time series anomaly detection with association discrepancy. In. ICLR. https://openreview.net/pdf?id=LzQQ89U1qm_
Yang, B., Long, W., Zhang, Y., Xi, Z., Jiao, J., & Li, Y. (2024). Multivariate time series anomaly detection: Missing data handling and feature collaborative analysis in robot joint data. Journal of Manufacturing Systems, 75, 132-149. https://doi.org/10.1016/j.jmsy.2024.06.006
Yang, L., Yan, W., & Wu, H. (2021). Comparison of deep learning-based methods in multimodal anomaly detection: a case study in human–robot collaboration. Science Progress, 104(2), 00368504211021192. https://doi.org/10.1177/00368504211021192
Yun, H., Kim, H., Jeong, Y. H., & Jun, M. B. (2023). Autoencoder-based anomaly detection of industrial robot arm using stethoscope based internal sound sensor. Journal of Intelligent Manufacturing, 34(3), 1427-1444. https://doi.org/10.1007/s10845-021-01862-4
Zhong, Z., Zhao, Y., Yang, A., Zhang, H., Qiao, D., & Zhang, Z. (2022). Industrial Robot Vibration Anomaly Detection Based on Sliding Window One-Dimensional Convolution Autoencoder. Shock and Vibration, 2022(1), 1179192. https://doi.org/10.1155/2022/1179192
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Reports in Mechanical Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.