Detecting Contextual Faults in Unmanned Aerial Vehicles Using Dynamic Linear Regression and K-Nearest Neighbour Classifier
https://doi.org/10.17285/0869-7035.0024
Abstract
Unmanned aerial vehicle (UAV) is a complex system. Its design involves control, aerodynamics, and communication systems. We use the complex linear relationships among UAV attributes (sensor readings, and commands) to propose a new technique to detect contextual faults. The contextual faults mean that a defective sensor shows invalid values concerning the context of other attributes. The proposed approach depends on estimating the values of a focused attribute using dynamic linear regression. Next, it calculates the estimation error at each time step. The values of the estimation error are classified using K-NN (Nearest Neighbour) classifier into two classes (Normal, Abnormal). The abnormal points are flagged as potential faults. Moreover, comparison with other algorithms (K-Means and One-Class SVM) is made. The proposed approach showed better results in most of the cases.
About the Authors
A. Alos
Department of Informatics, Higher Institute for Applied Sciences and Technology, Damascus, Syria
Syrian Arab Republic
Syrian Arab Republic
Z. Dahrouj
Department of Informatics, Higher Institute for Applied Sciences and Technology, Damascus, Syria
Russian Federation
Russian Federation
References
1. Colomina, I., Molina, P., Unmanned aerial systems for photogrammetry and remote sensing: A review, ISPRS Journal of Photogrammetry and Remote Sensing, 2014, vol. 92, pp. 79–97. DOI:10,1016/j. isprsjprs.2014.02.013.
2. Chandola, V., Banerjee, A., and Kumar, V., Anomaly detection: A survey, ACM Comput. Surv., 2009, vol. 41, pp. 1–58. DOI:10,1145/1541880,1541882.
3. Renckens, I., Automatic detection of suspicious behaviour, Master Thesis, 2014.
4. Sun, R., Cheng, Q., Wang, G., and Ochieng, W.Y., A novel online data-driven algorithm for detecting UAV navigation sensor faults, Sensors, 2017, vol. 17, no. 10, p. 2243. doi:10,3390/s17102243.
5. Ding, X., Li, Y., Belatreche, A., and Maguire, L.P., An experimental evaluation of novelty detection methods, Neurocomputing, 2014, vol. 135, pp. 313–327. DOI: 10,1016/j.neucom.2013.12.002.
6. Pasillas-Díaz, J.R., Ratté, S., An unsupervised approach for combining scores of outlier detection techniques, based on similarity measures, Electron. Notes Theor. Comput. Sci., 2016, vol. 329, pp. 61–77. DOI:10,1016/j.entcs.2016.12.005.
7. Khalastchi, E., Kalech, M., Kaminka, G.A., and Lin, R., Online data-driven anomaly detection in autonomous robots, Knowledge and Information Systems, 2015, vol. 43, pp. 657–688. DOI: 10,1007/ s10115-014-0754-y.
8. Cork, L., Walker, R., Sensor fault detection for UAVs using a nonlinear dynamic model and the IMMUKF algorithm, IEEE Information, Decision and Control, 2007, pp. 230–235.
9. Bu, J., Sun, R., Bai, H., Xu, R., Xie, F., Zhang, Y., and Ochieng, W.Y., Integrated method for the UAV navigation sensor anomaly detection, IET Radar, Sonar & Navigation, 2017, vol. 11, pp. 847–853. DOI:10,1049/iet-rsn.2016.0427.
10. Lin, R., Khalastchi, E., and Kaminka, G.A., Detecting anomalies in unmanned vehicles using the Mahalanobis distance, Proc. IEEE International Conference on Robotics and Automation, 2010, pp. 3038–3044. DOI:10,1109/ROBOT.2010,5509781.
11. Khalastchi, E., Kaminka, G.A., Kalech, M., and Lin, R., Online anomaly detection in unmanned vehicles, Proc. 10th International Conference on Autonomous Agents and Multiagent Systems, 2011, vol. 1, pp. 115–122.
12. Pokrajac, D., Latecki, L.J., and Lazarevic, A., Incremental local outlier detection for data streams, Proc. IEEE Symposium on Computational Intelligence and Data Mining, 2007, pp. 504–515. DOI:10,1109/CIDM.2007.368917. 13. Paffenroth, R., Kay, K., and Servi, L., Robust PCA for anomaly detection in cyber networks, ArXiv, 2018, https://arxiv.org/pdf/1801.01571.pdf.
13. Yong, D., Yaqing, X., Yuanpeng, Z., Yu, P., and Datong, L., Unmanned aerial vehicle sensor data anomaly detection using kernel principal component analysis, Proc. IEEE 13th International Conference on Electronic Measurement and Instruments, 2017, pp. 241–246.
14. Hundman, K., Constantinou, V., Laporte, C., Colwell, I., and Soderstrom, T., Detecting spacecraft anomalies using LSTMs and nonparametric dynamic thresholding, Proc. 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, New York, NY, 2018, pp. 387–395.
15. Weisberg, S., Applied Linear Regression, Hoboken, NJ: Wiley-Interscience, 2005.
16. Ullah, I., Fayaz, M., and Kim, D., Improving accuracy of the Kalman filter algorithm in dynamic conditions using ANN-based learning module, Symmetry, 2019, vol. 11, no. 1, p. 94. DOI:10,3390/ sym11010094.
17. Oza, N., FLTz flight simulator, https://c3.ndc.nasa.gov/dashlink/resources/294/
18. Chu, E., Gorinevsky, D., and Boyd, S.P., Detecting aircraft anomalies cruise flight data, Proc. AIAA Infotech Aerospace Conference, Atlanta, GA, 2010, pp. 1–15.
19. Le, V.-H., Kim, S.-R., K-strings algorithm, a new approach based on Kmeans, Proc. 2015 Conference on Research in Adaptive and Convergent Systems (RACS), 2015, pp. 15–20.
20. Ouyang, Q., Lu, W., Monthly rainfall forecasting using echo state networks coupled with data preprocessing methods, Water Resources Management, 2018, vol. 32, pp. 659–674. DOI:10,1007/ s11269-017-1832-1.
21. Karami, A., Guerrero-Zapata, M., A fuzzy anomaly detection system based on hybrid PSOKmeans algorithm in content-centric networks, Neurocomputing, 2015, vol. 149, pp. 1253–1269. DOI:10,1016/j.neucom.2014.08.070.
Review
For citations:
Alos A., Dahrouj Z. Detecting Contextual Faults in Unmanned Aerial Vehicles Using Dynamic Linear Regression and K-Nearest Neighbour Classifier. Giroskopiya i Navigatsiya. 2020;28(1):66-80. (In Russ.) https://doi.org/10.17285/0869-7035.0024
Views:
4
Email this article 