Abstract
The traditional fault diagnosis method of aircraft navigation system monitors the system integrity through redundant information, which has extremely high requirements on the accuracy of the model, and the gradual fault detection time is long, which cannot meet the real-time nature of aircraft navigation. In this paper, the minimum cost function corresponding to the threshold is constructed by using the probability of false alarms and the probability of missed failures in the system, and the threshold of the system is obtained by the minimum value of the cost function. In addition, the actual navigation data is preprocessed, and the corresponding labels are sent to the convolutional neural network for training to obtain a fault diagnosis model. Using One-Class SVM to train normal navigation data, the fault diagnosis of the navigation system is realized. Using the above three methods, a comprehensive discriminant system is constructed to diagnose the faults of the navigation system. Experiments show that the detection time of the comprehensive discrimination mechanism for slowly changing faults is significantly shorter than that of the threshold method, and the fault diagnosis accuracy rate reaches more than 97%.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Shahong, L.: Research on Strapdown Inertial Navigation System Multi-Sensor Redundancy Technology[J]. Harbin Engineering University, Harbin (2012)
Daly, K.C., Gai, E., Harrison, J.V.: Generalized likelihood test for FDI in redundant sensor configurations[J]. J. Guidance Control 2(1), 9–17 (1979)
Jin, H., Zhang, H.Y.: Optimal parity vector sensitive to designated sensor fault[J]. IEEE Trans. Aerosp. Electron. Syst. 35(4), 1122–1128 (1999)
Shim, D.S., Yang C.K.: Geometric FDI based on SVD for redundant inertial sensor systems[C]. In: 2004 5th Asian Control Conference (IEEE Cat. No. 04EX904). IEEE 2, pp. 1094-1100 (2004)
Gai, E.G., Adams, M.B., Walker, B.K.: Determination of failure thresholds in hybrid navigation[J]. IEEE Trans. Aerosp. Electron. Syst. 6, 744–755 (1976)
LeCun, Y., Bottou, L., Bengio, Y., et al.: Gradient-based learning applied to document recognition[J]. Proc. IEEE 86(11), 2278–2324 (1998)
Krizhevsky, A., Sutskever, I., Hinton,G.E.: ImageNet classification with deep convolutional neural networks[J]. Adv. Neural Inf. Process. Syst. 25, (2012)
Ruff L, Vandermeulen R, Goernitz N, et al. Deep One-Class Classification[C]. In: PMLR, pp. 4393-4402 (2018)
Noble, W.S.: What is a support vector machine?[J]. Nat. Biotechnol. 24(12), 1565–1567 (2006)
Suthaharan, S.: Support Vector Machine Learning Models and Algorithms for Big Data Classification, pp. 207–235. Springer, Boston, MA (2016). https://doi.org/10.1007/978-1-4899-7641-3
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Mao, Y., Deng, S., Chen, D., Qin, N., Huang, D. (2023). Fault Diagnosis of Aircraft Navigation System Based on Machine Learning. In: Yan, L., Duan, H., Deng, Y. (eds) Advances in Guidance, Navigation and Control. ICGNC 2022. Lecture Notes in Electrical Engineering, vol 845. Springer, Singapore. https://doi.org/10.1007/978-981-19-6613-2_144
Download citation
DOI: https://doi.org/10.1007/978-981-19-6613-2_144
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-6612-5
Online ISBN: 978-981-19-6613-2
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)