Hybrid AI/ML Framework for Real-time Cybersecurity in Unmanned Aerial Systems: Detection of GPS Spoofing and Command Injection Attacks
Keywords:
Artificial Intelligence (AI), Cybersecurity, Intrusion Detection Systems (IDS), Machine Learning (ML), Unmanned Aerial Systems (UAS)Abstract
The increasing reliance on Unmanned Aerial Systems (UAS) for critical tasks has highlighted the pressing need for robust cybersecurity measures to safeguard against evolving threats, particularly GPS spoofing and command injection attacks. Despite extensive research in drone cybersecurity, existing studies often utilize simulated data or isolated sensor analysis, lacking practical applicability and comprehensive, real-world validation. Addressing this gap, this research proposes an adaptive Artificial Intelligence and Machine Learning (AI/ML) cybersecurity framework specifically tailored for real-time intrusion detection and threat classification in UAS operations. This paper developed an innovative cybersecurity testbed by integrating a DJI drone with an ECSTAR-developed companion computer, facilitating the collection of extensive and realistic flight telemetry data (approximately 50,000 samples) under normal and intentionally simulated cyber-attack scenarios. Original contribution of this research lies in the design and deployment of a hybrid detection model that combines an autoencoder for efficient anomaly detection with an LSTM network for accurate temporal classification of cyber threats. Comparative evaluations demonstrate that our model significantly improves upon traditional detection techniques by achieving superior accuracy (approximately 95%) and rapid detection latency suitable for onboard real-time implementation. Moreover, incorporating adaptive thresholding and robust fallback mechanisms further enhances the practical resilience of the solution, representing a significant advancement over existing approaches. This research thus contributes a validated, real-time capable cybersecurity framework, along with a valuable benchmark dataset, providing a robust foundation for future advancements in autonomous drone cybersecurity.
References
E. Marchetti, T. Waheed, and A. Calabrò, “Cybersecurity testing in drones domain: A systematic literature review,” in IEEE Access, vol. 12, pp. 171166–171184, 2024, doi: https://doi.org/10.1109/access.2024.3495994
A. Baron Garcia, “Machine Learning and Artificial Intelligence Methods for Cybersecurity Data within the Aviation Ecosystem,” Doctoral Dissertations and Master’s Theses, Oct. 2022, Available: https://commons.erau.edu/edt/700/
B. Pardhasaradhi and L. R. Cenkeramaddi, “GPS spoofing detection and mitigation for drones using distributed radar tracking and fusion,” in IEEE Sensors Journal, vol. 22, no. 11, pp. 11122–11134, June, 2022, doi: https://doi.org/10.1109/jsen.2022.3168940
T. D. Lofstrom, “A methodology of hacking domestic unmanned aerial systems,” M.S. thesis, Dept. Mech. Eng., Univ. of North Dakota, Grand Forks, ND, USA, 2022.
M. McNabb, “Clarifying cybersecurity guidelines for drones: The DETECT Act,” Dronelife, Mar. 27, 2024. Available: https://dronelife.com/2024/03/27/clarifying-cybersecurity-guidelines-for-drones-the-detect-act/
G. P. Rimoli, F. Palmieri, and M. Ficco, “Synthetic threat dataset generation by UAV fleet simulation,” in Proceedings of the 14th International Defense and Homeland Security Simulation Workshop (DHSS 2024), Tenerife, Spain, Sep. 2024, doi: https://doi.org/10.46354/i3m.2024.dhss.002
M. Nayfeh, J. Price, M. Alkhatib, K. Al Shamaileh, N. Kaabouch and V. Devabhaktuni, “A real-time machine learning-based GPS spoofing solution for location-dependent UAV applications,” 2023 IEEE International Conference on Electro Information Technology (eIT), Romeoville, IL, USA, 2023, pp. 289–293, doi: https://doi.org/10.1109/eit57321.2023.10187344
G. Aissou, H. O. Slimane, S. Benouadah and N. Kaabouch, “Tree-based supervised machine learning models for detecting GPS spoofing attacks on UAS,” 2021 IEEE 12th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), New York, NY, USA, 2021, pp. 0649–0653, doi: https://doi.org/10.1109/uemcon53757.2021.9666744
S. Jeong, E. Park, K. Uk Seo, J. D. Yoo, and H. K. Kim, “MUVIDS: False MAVLink injection attack detection in communication for unmanned vehicles,” Workshop on Automotive and Autonomous Vehicle Security (AutoSec) 2021, Feb. 2021, doi: https://doi.org/10.14722/autosec.2021.23036
MAVLink, “MAVLink developer guide,” Mavlink, 2025. Available: https://mavlink.io/en/
Y. Özkan, “Detection of attacks in network traffic with the autoencoder-based unsupervised learning method,” Acta Infologica, vol. 6, no. 2, pp. 199–207, 2022, doi: https://doi.org/10.26650/acin.1142806
H. Hacılar, B. K. Dedeturk, B. Bakir-Gungor, and V. C. Gungor, “Network anomaly detection using deep autoencoder and parallel artificial bee colony algorithm-trained neural network,” PeerJ Computer Science, Oct. 2024, doi: https://doi.org/10.7717/peerj-cs.2333
Z. Que, Y. Liu, C. Guo, X. Niu, Y. Zhu and W. Luk, “Real-time anomaly detection for flight testing using autoencoder and LSTM,” 2019 International Conference on Field-Programmable Technology (ICFPT), Tianjin, China, 2019, pp. 379–382, doi: https://doi.org/10.1109/icfpt47387.2019.00072
R. Shrestha et al., “Anomaly detection based on LSTM and autoencoders using federated learning in smart electric grid,” Journal of parallel and distributed computing, vol. 193, Nov. 2024, doi: https://doi.org/10.1016/j.jpdc.2024.104951
K. Amancio, G. Pinheiro, L. Teixeira, and C. Duarte, “Enabling intelligent and autonomous drones using embedded Linux,” in 10th International Micro-Air Vehicles Conference, 2018. Available: https://www.imavs.org/papers/2018/44.pdf
A. S. Mekam, “Performance analysis of the communication system of a drone prototype used for maintenance and cleaning,” M.S. thesis, Dept. Telecommun. Syst., Blekinge Inst. Technol., Karlskrona, Sweden, Feb. 2023. Available: https://www.diva-portal.org/smash/get/diva2:1740985/FULLTEXT02.pdf%E2%80%8B
A. Flores, “Autonomous drone communication using LoRa and MAVLink for long-range control,” M.S. thesis, Dept. Comput. Eng., California State Univ., Northridge, CA, USA, Dec. 2024. Available: https://scholarworks.calstate.edu/downloads/3n2047453
M. Calvo-Fullana, D. Mox, A. Pyattaev, J. Fink, V. Kumar and A. Ribeiro, “ROS-NetSim: A framework for the integration of robotic and network simulators,” in IEEE Robotics and Automation Letters, vol. 6, no. 2, pp. 1120–1127, April 2021, doi: https://doi.org/10.1109/lra.2021.3056347
TensorFlow, “An end-to-end platform for machine learning,” TensorFlow, 2025. Available: https://www.tensorflow.org
A. Oluwasanmi, M. U. Aftab, E. Baagyere, Z. Qin, M. Ahmad, and M. Mazzara, “Attention autoencoder for generative latent representational learning in anomaly detection,” Sensors, vol. 22, no. 1, Dec. 2021, doi: https://doi.org/10.3390/s22010123
A. Krizhevsky, I. Sutskever, and G. E. Hinton, “ImageNet classification with deep convolutional neural networks,” Communications of the ACM, vol. 60, no. 6, pp. 84–90, May 2017, doi: https://doi.org/10.1145/3065386
L. Huang, X. Pan, Y. Liu, and L. Gong, “An unsupervised machine learning approach for monitoring data fusion and health indicator construction,” Sensors, vol. 23, no. 16, Aug. 2023, doi: https://doi.org/10.3390/s23167239
S. Teng, Z. Liu, W. Luo, G. Chen, and L. Cheng, “Bridge anomaly detection based on reconstruction error and structural similarity of unsupervised convolutional auto-encoder,” Structural Health Monitoring, vol. 23, no. 4, pp. 2221–2237, Nov. 2023, doi: https://doi.org/10.1177/14759217231200096
İ. Üstek, M. Arana-Catania, A. Farr, and I. Petrunin, “Deep autoencoders for unsupervised anomaly detection in wildfire prediction,” Earth and Space Science, vol. 11, no. 11, Nov. 2024, doi: https://doi.org/10.1029/2024EA003997
A. A. Attar, K. Bao, V. Hagenmeyer, T. Fabarisov and A. Morozov, “Improving anomaly detection with adaptive dynamic threshold: A review and enhanced method,” 2024 8th International Conference on System Reliability and Safety (ICSRS), Sicily, Italy, 2024, pp. 662–666, Available: https://ieeexplore.ieee.org/document/10927575
A. Mallick, K. Hsieh, B. Arzani, and G. Joshi, “Matchmaker: Data drift mitigation in machine learning for large-scale systems,” in Proceedings of the 5th Workshop on Machine Learning and Systems, Santa Clara, CA, USA, 2022, Available: https://proceedings.mlsys.org/paper_files/paper/2022/file/069a002768bcb31509d4901961f23b3c-Paper.pdf
F. T. Liu, K. M. Ting and Z. -H. Zhou, “Isolation forest,” 2008 Eighth IEEE International Conference on Data Mining, Pisa, Italy, 2008, pp. 413–422, Available: https://ieeexplore.ieee.org/document/4781136
H. H. Nguyen, C. N. Nguyen, X. T. Dao, Q. T. Duong, D. P. T. Kim, and M.-T. Pham, “Variational autoencoder for anomaly detection: A comparative study,” arXiv preprint arXiv:2408.13561, Aug. 2024, Available: https://arxiv.org/html/2408.13561v1
W. Wei, J. Banerjee, S. Islam, C. Pan and M. Xie, “Energy-aware incremental ota update for flash-based battery less IoT devices,” 2024 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), Knoxville, TN, USA, 2024, pp. 51–56, doi: https://doi.org/10.1109/ISVLSI61997.2024.00021
Q. Zhou et al., “Shale oil production predication based on an empirical model-constrained CNN–LSTM,” Energy Geoscience, vol. 5, no. 2, Apr. 2024, doi: https://doi.org/10.1016/j.engeos.2023.100252
B. M. Hussein and S. M. Shareef, “An empirical study on the correlation between early stopping patience and epochs in deep learning,” 2nd International Conference on Applied Computing & Smart Cities (ICACS24), vol. 64, 2024, doi: https://doi.org/10.1051/itmconf/20246401003
O. Jullian, B. Otero, M. Stojilović, J. J. Costa, J. Verdú, and M. A. Pajuelo, “Deep learning detection of GPS spoofing,” in Machine Learning, Optimization, and Data Science: 7th International Conference, LOD 2021, Grasmere, UK, 2022, pp. 527–540, doi: https://doi.org/10.1007/978-3-030-95467-3_38
S. Gou et al., “Dynamic spatio-temporal pruning for efficient spiking neural networks,” Frontiers in Neuroscience, vol. 19, Mar. 2025, doi: https://doi.org/10.3389/fnins.2025.1545583
X. Xu and W. Ren, “A hybrid model of stacked autoencoder and modified particle swarm optimization for multivariate chaotic time series forecasting,” Applied Soft Computing, vol. 116, Feb. 2022, doi: https://doi.org/10.1016/j.asoc.2021.108321
M. Gamal, M. Elhamahmy, S. Taha, and H. Elmahdy, “Improving intrusion detection using LSTM–RNN to protect drones’ networks,” Egyptian Informatics Journal, vol. 27, Sept. 2024, doi: https://doi.org/10.1016/j.eij.2024.100501
M. Rashidi, “Application of TensorFlow lite on embedded devices : A hands-on practice of TensorFlow model conversion to TensorFlow Lite model and its deployment on Smartphone to compare model’s performance,” DIVA, 2022. https://www.diva-portal.org/smash/record.jsf?dswid=-5568&pid=diva2%3A1698946
T. Guo, Z. Xu, X. Yao, H. Chen, K. Aberer and K. Funaya, “Robust online time series prediction with recurrent neural networks,” 2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA), Montreal, QC, Canada, 2016, pp. 816–825, Available: https://ieeexplore.ieee.org/document/7796970
D. L. Vajda, T. V. Do, T. Bérczes, and K. Farkas, “Machine learning-based real-time anomaly detection using data pre-processing in the telemetry of server farms,” Scientific Reports, vol. 14, no. 1, Oct. 2024, doi: https://doi.org/10.1038/s41598-024-72982-z
Y.-B. Bae, Y.-U. Kim, J.-O. Park, and H.-S. Ahn, “Autonomous cooperative levels of multiple-heterogeneous unmanned vehicle systems,” arXiv Preprint, arXiv:2405.09193, May 2024, Available: https://arxiv.org/pdf/2405.09193
Decent Cybersecurity, “Real-time anomaly detection in drone IFF responses: Enhancing UAV security with AI,” Decent Cybersecurity, Aug. 24, 2024, Available: https://decentcybersecurity.eu/real-time-anomaly-detection-in-drone-iff-responses-enhancing-uav-security-with-ai/
I. Fosić, D. Žagar, K. Grgić, and V. Križanović, “Anomaly detection in NetFlow network traffic using supervised machine learning algorithms,” Journal of Industrial Information Integration, vol. 33, Jun. 2023, doi: https://doi.org/10.1016/j.jii.2023.100466
M. C. Vattikuti, “Real-time anomaly detection in industrial IoT systems using hybrid AI models,” International Scientific Journal for Research, vol. 5, no. 5, 2023, Available: https://isjr.co.in/index.php/ISJR/article/view/286.
M.-A. Mendoza and H. R. Amistadi, “Machine learning for anomaly detection on VM and host performance metrics,” MITRE Corporation, Bedford, MA, 2018. Available: https://www.mitre.org/sites/default/files/publications/pr-18-0941-machine-learning-anomaly-detection.pdf
M. S. Lakshmi, G. Rajavikram, V. Dattatreya, B. S. Jyothi, S. Patil, and M. Bhavsingh, “Evaluating the isolation forest method for anomaly detection in software defined networking security,” Journal of Electrical Systems, vol. 19, no. 4, pp. 279–297, 2023, doi: https://doi.org/10.52783/jes.639
M. Eriksson and A. Johansson, “Narrow pretraining of deep neural networks: Exploring autoencoder pretraining for anomaly detection on limited datasets in non-natural image domains,” M.S. thesis, Dept. Sci. Technol., Linköping Univ., Norrköping, Sweden, 2022. Available:https://www.diva-portal.org/smash/get/diva2:1686601/FULLTEXT01.pdf%E2%80%8B
