LANDMINE RECOGNITION USING A KOLMOGOROV–ARNOLD NEURAL NETWORK BASED ON MAGNETIC SENSING DATA

Ivan Peleshchak, Viacheslav Beltiukov

Abstract


Background. Landmine contamination remains a critical issue for more than 60 countries, including Ukraine, where the recovery of agriculture and infrastructure is hampered by hidden explosive devices. The authors propose a passive land-mine recognition approach that combines measurements of magnetic anomalies obtained with an FLC-100 sensor and a Kolmogorov-Arnold Network (KAN), delivering high accuracy while minimising the risk of detonation.

Materials and Methods. A baseline data set of 338 real measurements (sensor voltage V, sensor height H, soil type S) was balanced by generating 50 synthetic records for every “soil–mine” pair using a parameterised normal distribution. After normalising V and applying one-hot encoding to S and the mine classes M, a three-dimensional feature space was formed. Two KAN architectures were evaluated: KAN (3-16-16-4) and KAN (3-64-64-4), both employing cubic B-splines to achieve high-precision mine recognition (> 95 % accuracy). Training was conducted in PyKAN (PyTorch backend) for 65 epochs with a fixed spline grid (k = 3, m = 10) using the Adam optimiser.

Results and Discussion. The KAN (3-16-16-4) model achieved an accuracy of 93.56 % without overfitting; the main confusion occurred between the “anti-personnel” and “booby-trap” classes. Increasing the number of neurons in each hidden layer to 64 raised the accuracy to 95.59 % and eliminated the erroneous assignment of “anti-personnel” mines to the “booby-trap” class. Both networks perfectly distinguished the “no-mine” and “anti-tank” cases, confirming the robustness of spline activations to sensor noise.

Conclusion. The computer experiment shows that a Kolmogorov–Arnold neural network with cubic B-spline weights provides robust recognition of different mine types (“no mine”, “anti-tank mine”, “anti-personnel mine”, “booby-trap”) using magnetic-field sensor data (10⁻¹⁰ Tesla) with accuracy exceeding 95 %. Interpretable spline weights allow the contribution of each feature to be analysed, ensuring high sensitivity to small anomalies and demonstrating the scalability of KAN.

Keywords: Kolmogorov–Arnold network, fluxgate magnetic sensor, passive mine detection


Full Text:

PDF

References


  1. Landmine Monitor 2023. (2023). https://backend.icblcmc.org/assets/reports/Landmine-Monitors/LMM2023/Downloads/Landmine-Monitor-2023_web.pdf
  2. Lytvyn, V., Peleshchak, I., Peleshchak, R., Mediakov, O., & Pukach, P. (2023). Development of a hybrid neural network model for mine detection by using ultrawideband radar data. Eastern-European Journal of Enterprise Technologies, 3(9 (123)), 78–85. https://doi.org/10.15587/1729-4061.2023.279891
  3. Peleshchak, R. M., Lytvyn, V. V., Nazarkevych, M. A., Peleshchak, I. R., & Nazarkevych, H. Y. (2024). Influence of the symmetry neural network morphology on the mine detection metric. Symmetry, 16(4), 485. https://doi.org/10.3390/sym16040485
  4. Yilmaz, C., Kahraman, H. T., & Soyler, S. (2018). Passive mine detection and Classification Method based on hybrid model. IEEE Access, 6, 47870–47888. https://doi.org/10.1109/access.2018.2866538
  5. Liu, Z., Wang, Y., Vaidya, S., Ruehle, F., Halverson, J., Soljačić, M., Hou, T. Y., & Tegmark, M. (2024). KAN: Kolmogorov-Arnold Networks. arXiv. https://doi.org/10.48550/arxiv.2404.19756
  6. Erdmann, J., Mausolf, F., & Späh, J. L. (2024). KAN we improve on HEP classification tasks? Kolmogorov-Arnold Networks applied to an LHC physics example. arXiv. https://doi.org/10.48550/arxiv.2408.02743
  7. Somvanshi, S., Javed, S. A., Islam, M. M., Pandit, D., & Das, S. (2024). A survey on Kolmogorov-Arnold Network. arXiv. https://doi.org/10.48550/arxiv.2411.06078
  8. Barašin, I., Bertalanič, B., Mohorčič, M., & Fortuna, C. (2024). Exploring Kolmogorov-Arnold networks for Interpretable Time Series Classification. arXiv. https://doi.org/10.48550/arxiv.2411.14904
  9. Wang, S., Yue, X., Lyu, Y., & Shan, C. (2025). Exploring Adversarial Transferability between Kolmogorov-arnold Networks. arXiv.org. https://doi.org/10.48550/arXiv.2503.06276
  10. Cheon, M. (2024). Kolmogorov-Arnold Network for Satellite image classification in remote sensing. arXiv. https://doi.org/10.48550/arxiv.2406.00600
  11. Drokin, I. (2024). Kolmogorov-Arnold Convolutions: Design Principles and Empirical Studies. arXiv. https://doi.org/10.48550/arxiv.2407.01092
  12. Behzad. (2018). Data Sheet FLC 100. Scribd. https://scribd.com/document/382854526/Data-Sheet-FLC-100
  13. Maharana, K., Mondal, S., & Nemade, B. (2022). A review: Data pre-processing and data augmentation techniques. Global Transitions Proceedings, 3(1), 91–99. https://doi.org/10.1016/j.gltp.2022.04.020
  14. Qamar, R., & Zardari, B. A. (2023). Artificial Neural Networks: An Overview. Mesopotamian Journal of Computer Science, 130–139. https://doi.org/10.58496/mjcsc/2023/015
  15. Vanneschi, L., & Silva, S. (2023). Support vector machines. In Natural computing series, 271–281. https://doi.org/10.1007/978-3-031-17922-8_10
  16. Lampinen, J., & Vehtari, A. (2001). Bayesian approach for neural networks—review and case studies. Neural Networks, 14(3), 257–274. https://doi.org/10.1016/s0893-6080(00)00098-8
  17. De Ville, B. (2013). Decision trees. Wiley Interdisciplinary Reviews Computational Statistics, 5(6), 448–455. https://doi.org/10.1002/wics.1278
  18. Schmidt-Hieber, J. (2020). The Kolmogorov-Arnold representation theorem revisited. arXiv. https://doi.org/10.48550/arxiv.2007.15884
  19. Chaudhuri, A. (2021). B-Splines. arXiv. https://doi.org/10.48550/arxiv.2108.06617
  20. KindXiaoming. (2024). GitHub - KindXiaoming/pykan: Kolmogorov Arnold Networks. GitHub. https://github.com/KindXiaoming/pykan
  21. Eberhart, R. C., Dobbins, R. W., & Hutton, L. V. (1990). Performance Metrics. In Elsevier eBooks, 161–176. https://doi.org/10.1016/b978-0-12-228640-7.50013-1




DOI: http://dx.doi.org/10.30970/eli.30.10

Refbacks

  • There are currently no refbacks.