Dynamic trend perception for fine-grained unmanned aerial vehicle anomaly detection in flight data

Abstract

With the widespread application of UAVs, ensuring their safety and reliability is increasingly crucial. Data-driven anomaly detection methods demonstrate outstanding performance in unmanned aerial vehicle (UAV) health monitoring. Previous studies extract spatiotemporal features from high-dimensional flight data, but emphasize intricate multidimensional correlations, neglecting the intrinsic characteristics of single-dimensional data. This oversight leads to insufficient supervision of intrinsic univariate information. Consequently, subtle deviations in individual flight variables may be masked by dominant multidimensionality, limiting the model's ability to distinguish fine-grained anomalies from normal flight states. To address this problem, we propose F-UAD, a novel dynamic trend perception framework for fine-grained UAV anomaly detection. F-UAD is designed to coordinate univariate feature preservation, multidimensional coupling extraction, and dynamic trend perception, enabling fine-grained anomaly detection of subtle deviations in UAV flight data. During training, the dynamic trend perception module incorporates (1) multiscale spatiotemporal feature reconstruction to balance independent univariate feature preservation with comprehensive multidimensional coupling extraction and (2) dynamic trend forecasting based on decomposed flight trends to improve the perception of complex temporal evolution. During testing, reconstruction consistency and dynamic trend consistency are jointly used to compute anomaly scores, thereby enhancing the separation between normal and abnormal flight states. Experiments demonstrate that our F-UAD framework outperforms state-of-the-art methods across six AirLab failure and anomaly (ALFA) real flight dataset subsets, improving UAV reliability.

Keywords

Anomaly detection / Multivariate time series / Unsupervised anomaly detection / Unmanned aerial vehicle (UAV) / Flight data

References

[1] Ahmad MW , Akram MU , Ahmad R , et al., 2022. Intelligent framework for automated failure prediction, detection, and classification of mission critical autonomous flights. ISA Trans, 129: 355- 371.

[2] An J , Cho S , 2015. Variational autoencoder based anomaly detection using reconstruction probability. Spec Lect IE, 2 (1): 1- 18.

[3] Audibert J , Michiardi P , Guyard F , et al., 2020. USAD:unsupervised anomaly detection on multivariate time series. Proc 26^th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.3395- 3404.

[4] Baskaya E , Bronz M , Delahaye D , 2017. Fault detection & diagnosis for small UAVs via machine learning. Proc IEEE/AIAA 36^th Digital Avionics Systems Conf, p.1- 6.

[5] Bell V , Rengasamy D , Rothwell B , et al., 2022. Anomaly detection for unmanned aerial vehicle sensor data using a stacked recurrent autoencoder method with dynamic thresholding. https://arxiv.org/abs/2203.04734

[6] Bontemps L , Cao VL , McDermott J , et al., 2016. Collective anomaly detection based on long short-term memory recurrent neural networks. Proc 3^rd Int Conf on Future Data and Security Engineering, p.141- 152.

[7] Braei M , Wagner S , 2020. Anomaly detection in univariate time-series:a survey on the state-of-the-art. https://arxiv.org/abs/2004.00433

[8] Chandola V , 2009. Anomaly Detection for Symbolic Sequences and Time Series Data. PhD Dissertation, University of Minnesota, Minneapolis, USA.

[9] Chen MZ , Zhang XC , Xiong XS , et al., 2021. Transformer:a multifunctional fast unmanned aerial vehicles-unmanned surface vehicles coupling system. Machines, 9 (8): 146.

[10] Chen YF , Wang BK , Liu W , et al., 2017. On-line and non-invasive anomaly detection system for unmanned aerial vehicle. Proc Prognostics and System Health Management Conf, p.1- 7.

[11] Choi K , Yi JH , Park C , et al., 2021. Deep learning for anomaly detection in time-series data:review, analysis, and guidelines. IEEE Access, 9: 120043- 120065.

[12] D'Amato E , Mattei M , Notaro I , et al., 2018. UAV sensor FDI in duplex attitude estimation architectures using a set-based approach. IEEE Trans Instrum Meas, 67 (10): 2465- 2475.

[13] Darban ZZ , Webb GI , Pan SR , et al., 2025a. CARLA:self-supervised contrastive representation learning for time series anomaly detection. Patt Recogn, 157: 110874.

[14] Darban ZZ , Webb GI , Pan SR , et al., 2025b. Deep learning for time series anomaly detection:a survey. ACM Comput Surv, 57 (1): 15.

[15] Deng AL , Hooi B , 2021. Graph neural network-based anomaly detection in multivariate time series. Proc 35^th AAAI Conf on Artificial Intelligence, p.4027- 4035.

[16] Garg A , Zhang WY , Samaran J , et al., 2022. An evaluation of anomaly detection and diagnosis in multivariate time series. IEEE Trans Neur Netw Learn Syst, 33 (6): 2508- 2517.

[17] Guo DF , Zhong MY , Zhou DH , 2018. Multisensor data-fusion-based approach to airspeed measurement fault detection for unmanned aerial vehicles. IEEE Trans Instrum Meas, 67 (2): 317- 327.

[18] Harrou F , Sun Y , Kadri F , et al., 2015. Early detection of abnormal patient arrivals at hospital emergency department. Proc Int Conf on Industrial Engineering and Systems Management, p.221- 227.

[19] Homayouni H , Ray I , Ghosh S , et al., 2021. Anomaly detection in COVID-19 time-series data. SN Comput Sci, 2 (4): 279.

[20] Jiang GQ , Nan PC , Zhang JC , et al., 2024. Robust spatial-temporal autoencoder for unsupervised anomaly detection of unmanned aerial vehicle with flight data. IEEE Trans Instrum Meas, 73: 3526014.

[21] Keipour A , Mousaei M , Scherer S , 2019. Automatic real-time anomaly detection for autonomous aerial vehicles. Proc Int Conf on Robotics and Automation, p.5679- 5685.

[22] Khosravi MR , Samadi S , 2021. BL-ALM:a blind scalable edge-guided reconstruction filter for smart environmental monitoring through green IoMT-UAV networks. IEEE Trans Green Commun Netw, 5 (2): 727- 736.

[23] Li D , Chen DC , Jin BH , et al., 2019. MAD-GAN:multivariate anomaly detection for time series data with generative adversarial networks. Proc 28^th Int Conf on Artificial Neural Networks, p.703- 716.

[24] Lin S , Clark R , Birke R , et al., 2020. Anomaly detection for time series using VAE-LSTM hybrid model. Proc IEEE Int Conf on Acoustics, Speech and Signal Processing, p.4322- 4326.

[25] Liu DT , Pang JY , Song G , et al., 2017. Fragment anomaly detection with prediction and statistical analysis for satellite telemetry. IEEE Access, 5: 19269- 19281.

[26] Liu DT , Song YC , Li L , et al., 2018. Online life cycle health assessment for lithium-ion battery in electric vehicles. J Cleaner Prod, 199: 1050- 1065.

[27] Liu PF , Sun XM , Han Y , et al., 2022. Arrhythmia classification of LSTM autoencoder based on time series anomaly detection. Biomed Signal Process Contr, 71: 103228.

[28] Maddikunta PKR , Hakak S , Alazab M , et al., 2021. Unmanned aerial vehicles in smart agriculture:applications, requirements, and challenges. IEEE Sens J, 21 (16): 17608- 17619.

[29] Marcos A , Waitman S , Sato M , 2022. Fault tolerant linear parameter varying flight control design, verification and validation. J Franklin Inst, 359 (2): 653- 676.

[30] Nam Y , Yoon S , Shin Y , et al., 2024. Breaking the time-frequency granularity discrepancy in time-series anomaly detection. Proc ACM Web Conf, p.4204- 4215.

[31] Narasimhan S , Brownston L , 2007. HyDE:a general framework for stochastic and hybrid model-based diagnosis. Proc 18^th Int Workshop on Principles of Diagnosis, p.186- 193.

[32] Nguyen HD , Tran KP , Thomassey S , et al., 2021. Forecasting and anomaly detection approaches using LSTM and LSTM autoencoder techniques with the applications in supply chain management. Int J Inform Manag, 57: 102282.

[33] Oussous A , Benjelloun FZ , Ait Lahcen A , et al., 2018. Big data technologies:a survey. J King Saud Univ-Comput Inform Sci, 30 (4): 431- 448.

[34] Pang JY , Liu DT , Peng Y , et al., 2015. Detecting continual anomalies in monitoring data stream based on sampling GPR algorithm. Proc IEEE Conf on Prognostics and Health Management, p.1- 7.

[35] Park JH , Shanbhag S , Chang DE , 2022. Model-free unsupervised anomaly detection of a general robotic system using a stacked LSTM and its application to a fixed-wing unmanned aerial vehicle. Proc IEEE/RSJ Int Conf on Intelligent Robots and Systems, p.4287- 4293.

[36] Park KH , Park EJ , Kim HK , 2021. Unsupervised fault detection on unmanned aerial vehicles:encoding and thresholding approach. Sensors, 21 (6): 2208.

[37] Qin HM , Yan MR , Ji HJ , 2021. Application of controller area network(CAN) bus anomaly detection based on time series prediction. Veh Commun, 27: 100291.

[38] Radoglou-Grammatikis P , Sarigiannidis P , Lagkas T , et al., 2020. A compilation of UAV applications for precision agriculture. Comput Netw, 172: 107148.

[39] Tuli S , Casale G , Jennings NR , 2022. TranAD:deep Transformer networks for anomaly detection in multivariate time series data. Proc VLDB Endowm, 15 (6): 1201- 1214.

[40] Ullah S , Kim KI , Kim KH , et al., 2019. UAV-enabled healthcare architecture:issues and challenges. Fut Gener Comput Syst, 97: 425- 432.

[41] Vaswani A , Shazeer N , Parmar N , et al., 2017. Attention is all you need. Proc 31^st Int Conf on Neural Information Processing Systems, p.6000- 6010.

[42] Wang T , Qiao MN , Lin ZW , et al., 2019. Generative neural networks for anomaly detection in crowded scenes. IEEE Trans Inform Forens Secur, 14 (5): 1390- 1399.

[43] Wu HX , HX JH , Wang JM , et al., 2021. Autoformer:decomposition Transformers with auto-correlation for long-term series forecasting. Proc 35^th Int Conf on Neural Information Processing Systems, Article 1717.

[44] Xiong WX , Wang P , Sun XC , et al., 2024. SiET:spatial information enhanced Transformer for multivariate time series anomaly detection. Knowl-Based Syst, 296: 111928.

[45] Yang L , Li SB , Li CJ , et al., 2023. A survey of unmanned aerial vehicle flight data anomaly detection:technologies, applications, and future directions. Sci China Technol Sci, 66 (4): 901- 919.

[46] Yang L , Li SB , Zhu CC , et al., 2024. Spatio-temporal correlationbased multiple regression for anomaly detection and recovery of unmanned aerial vehicle flight data. Adv Eng Inform, 60: 102440.

[47] Yang YY , Zhang CL , Zhou T , et al., 2023. DCdetector:dual attention contrastive representation learning for time series anomaly detection. Proc 29^th ACM SIGKDD Conf on Knowledge Discovery and Data Mining, p.3033- 3045.

[48] Yi YM , Zhang YM , 2017. Fault diagnosis of an unmanned quadrotor helicopter based on particle filter. Proc Int Conf on Unmanned Aircraft Systems, p.1432- 1437.

[49] Yin CY , Zhang S , Wang J , et al., 2022. Anomaly detection based on convolutional recurrent autoencoder for IoT time series. IEEE Trans Syst Man Cybern Syst, 52 (1): 112- 122.

[50] Zhang QC , Yang LT , Chen ZK , et al., 2018. A survey on deep learning for big data. Inform Fus, 42: 146- 157.

[51] Zhang YX , Chen YQ , Wang JD , et al., 2023. Unsupervised deep anomaly detection for multi-sensor time-series signals. IEEE Trans Knowl Data Eng, 35 (2): 2118- 2132.

[52] Zhao H , Wang YJ , Duan JY , et al., 2020. Multivariate time-series anomaly detection via graph attention network. Proc IEEE Int Conf on Data Mining, p.841- 850.

[53] Zhong J , Zhang YJ , Wang JY , et al., 2022. Unmanned aerial vehicle flight data anomaly detection and recovery prediction based on spatio-temporal correlation. IEEE Trans Rel, 71 (1): 457- 468.

[54] Zhou B , Liu SH , Hooi B , et al., 2019. BeatGAN:anomalous rhythm detection using adversarially generated time series. Proc 28^th Int Joint Conf on Artificial Intelligence, p.4433- 4439.

[55] Zhou X , Chu XY , Zou YQ , 2023. Sensor fault detection for UAVs based on MIC-LSTM with attention mechanism. Proc CAA Symp on Fault Detection, Supervision and Safety for Technical Processes, p.1- 6.

Rights & permissions

The Authors. Published by Zhejiang University Press Co., Ltd.