Hybrid Learning Model for Cross-Device Fault Detection of Industrial Robot Joints

doi:10.1007/s12204-025-2843-3

Abstract

Abstract: Industrial robots, widely employed to boost production efficiency, encounter escalating risks of joint faults as their service time lengthens. However, end-effector motion anomalies may stem from faults in the endeffector itself or from motion propagation in other joints. Moreover, the scarcity of fault samples for detection poses significant challenges. Install extra accelerometers for more precise fault diagnosis might increase the system’s complexity and costs. To tackle these challenges, this study leverages the ease of data acquisition to analyze current data from multi-joint industrial robots. A hybrid learning method is proposed for cross-device fault detection to identify the defective joint. This method integrates features from deep networks and spectral analysis to harness knowledge from both other robots and the target robot. An unsupervised model is used to assess the status of the joints based on the fused features. The proposed method’s effectiveness is validated through ablation studies and method comparisons. Results demonstrate that it accurately detects the abnormal joints without misjudgment.

Key words: fault detection, multi-joint industrial robot, hybrid learning, fused feature, current data

摘要： 工业机器人被广泛应用于提高生产效率，但随着其服役时间的延长，关节故障的风险也在不断增加。然而，末端执行器的运动异常可能源自末端执行器本身的故障，或者是由于其他关节的运动传播导致的。此外，用于故障检测的故障样本稀缺，这给故障诊断带来了巨大挑战。安装额外的加速度计以实现更精确的故障诊断可能会增加系统的复杂性和成本。为应对这些挑战，本研究利用数据采集的便利性，分析多关节工业机器人的电流数据。提出了一种用于跨设备故障检测的混合学习方法，以识别故障关节。该方法整合了深度网络和频谱分析的特征，从其他机器人和目标机器人中提取知识。采用基于融合特征的无监督模型来评估关节状态。通过消融实验和对比试验验证了所提方法的有效性。结果表明：该方法能够准确检测异常关节，且不会出现误判。

关键词: 故障检测，多关节工业机器人，混合学习方法，融合特征，电流数据

CLC Number:

TH-9

Xiao Lei, Zhao Hailong, Wu Xun, Wang Jun, Zhou Qihong. Hybrid Learning Model for Cross-Device Fault Detection of Industrial Robot Joints[J]. J Shanghai Jiaotong Univ Sci, 2026, 31(1): 82-98.

References

[1] JIANG P, ZHENG J J, WANG Z X, et al. Industrial robot energy consumption model identification: A coupling model-driven and data-driven paradigm [J]. Expert Systems with Applications, 2025, 262: 125604.
[2] LIN H I, MANDAL R, WIBOWO F S. BN-LSTM-based energy consumption modeling approach for an industrial robot manipulator [J]. Robotics and Computer-Integrated Manufacturing, 2024, 85: 102629.
[3] WEI X W, LI P C, TIAN W, et al. A fast dynamic pose estimation method for vision-based trajectory tracking control of industrial robots [J]. Measurement, 2024, 231: 114506.
[4] KIM Y, PARK J, NA K, et al. Phase-based time domain averaging (PTDA) for fault detection of a gearbox in an industrial robot using vibration signals [J]. Mechanical Systems and Signal Processing, 2020, 138: 106544.
[5] FU R, XIAO L, MA B T. Unsupervised fault detection of industrial robot joints using current signal [C]//2022 Global Reliability and Prognostics and Health Management. Yantai: IEEE, 2022: 1-8.
[6] CHEN C, LIU C, WANG T, et al. Compound fault diagnosis for industrial robots based on dual-transformer networks [J]. Journal of Manufacturing Systems, 2023, 66: 163-178.
[7] CHEN C, WANG T, LIU C, et al. Lightweight convolutional transformers enhanced meta-learning for compound fault diagnosis of industrial robot [J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 3520612.
[8] YANG Q B, LI X, WANG Y L, et al. Fault diagnosis of ball screw in industrial robots using non-stationary motor current signals [J]. Procedia Manufacturing, 2020, 48: 1102-1108.
[9] CHENG F Z, RAGHAVAN A, JUNG D, et al. High-accuracy unsupervised fault detection of industrial robots using current signal analysis [C]//2019 IEEE International Conference on Prognostics and Health Management. San Francisco: IEEE, 2019: 1-8.
[10] YUN H, KIM H, JEONG Y H, et al. Autoencoder-based anomaly detection of industrial robot arm using stethoscope based internal sound sensor [J]. Journal of Intelligent Manufacturing, 2023, 34(3): 1427-1444.
[11] WANG S J, TAO J F, JIANG Q C, et al. A digital twin framework for anomaly detection in industrial robot system based on multiple physics-informed hybrid convolutional autoencoder [J]. Journal of Manufacturing Systems, 2024, 77: 798-809.
[12] CHEN T T, LIU X P, XIA B Z, et al. Unsupervised anomaly detection of industrial robots using sliding-window convolutional variational autoencoder [J]. IEEE Access, 2020, 8: 47072-47081.
[13] XIA B J, WANG K, XU A D, et al. Intelligent fault diagnosis for bearings of industrial robot joints under varying working conditions based on deep adversarial domain adaptation [J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 3508313.
[14] OH Y, KIM Y, NA K, et al. A deep transferable motion-adaptive fault detection method for industrial robots using a residual–convolutional neural network [J]. ISA Transactions, 2022, 128: 521-534.
[15] YIN T, LU N, GUO G S, et al. Knowledge and data dual-driven transfer network for industrial robot fault diagnosis [J]. Mechanical Systems and Signal Processing, 2023, 182: 109597.
[16] ZHAO H L, XIAO L, ZHANG B Y. Cross-device anomaly detection of health status for industrial robot joints based on SWDCGAN-deep CORAL [C]//2023 Global Reliability and Prognostics and Health Management Conference. Hangzhou: IEEE, 2023: 1-8.
[17] KUMAR P, RAOUF I, KIM H S. Transfer learning for servomotor bearing fault detection in the industrial robot [J]. Advances in Engineering Software, 2024, 194: 103672.
[18] LONG J Y, MOU J D, ZHANG L W, et al. Attitude data-based deep hybrid learning architecture for intelligent fault diagnosis of multi-joint industrial robots [J]. Journal of Manufacturing Systems, 2021, 61: 736-745.
[19] HE Y M, ZHAO C, ZHOU X, et al. MJAR: A novel joint generalization-based diagnosis method for industrial robots with compound faults [J]. Robotics and Computer-Integrated Manufacturing, 2024, 86: 102668.
[20] LIU Y X, CHEN C, WANG T, et al. An attention enhanced dilated CNN approach for cross-axis industrial robotics fault diagnosis [J]. Autonomous Intelligent Systems, 2022, 2(1): 11.
[21] ZHONG Z D, ZHAO Y, YANG A Y, et al. Industrial robot vibration anomaly detection based on sliding window one-dimensional convolution autoencoder [J]. Shock and Vibration, 2022, 2022(1): 1179192.
[22] SUN B C, FENG J S, SAENKO K. Return of frustratingly easy domain adaptation [C]//Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence. Phoenix: ACM, 2016: 2058-2065.
[23] CHEN Y H, XIAO L, LI Z P. Partial domain fault diagnosis of bearings under cross-speed conditions based on 1D-CNN [C]//2022 Global Reliability and Prognostics and Health Management. Yantai: IEEE, 2022: 1-8.
[24] LI X Y, ZHANG Z, GAO L, et al. A new semi-supervised fault diagnosis method via deep CORAL and transfer component analysis [J]. IEEE Transactions on Emerging Topics in Computational Intelligence, 2022, 6(3): 690-699.