Технології виявлення вибухонебезпечних об’єктів із використанням штучного інтелекту

Vasyl Strutynskyi; Yuriy Danylchenko; Serhii Maidaniuk; Serhii Strutynskyi

doi:10.20535/2521-1943.2026.10.1(108).356762

Authors

Vasyl Strutynskyi National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Ukraine https://orcid.org/0000-0001-7167-0085
Yuriy Danylchenko National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Ukraine http://orcid.org/0000-0002-5375-950X
Serhii Maidaniuk National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Ukraine https://orcid.org/0000-0003-2853-8606
Serhii Strutynskyi National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, Ukraine https://orcid.org/0000-0001-9739-0399

DOI:

https://doi.org/10.20535/2521-1943.2026.10.1(108).356762

Keywords:

humanitarian demining, ground robotic systems, terrain scanning, explosive hazard detection and recognition, artificial intelligence, neural networks, fuzzy set theory, hyperspectral analysis

Abstract

Improving the safety of humanitarian demining during ground-based re-inspection of areas potentially contaminated with mines and explosive ammunition remains a critical challenge, particularly when mobile ground robotic systems are employed. Reliable detection and identification of such hazardous objects–often concealed, camouflaged, or occluded by foreign materials–require the integration of additional sensing methods with complementary or hybrid scanning techniques. However, this integration results in a substantial increase in data volume, necessitating the use of advanced processing approaches based on artificial intelligence. This paper proposes a technology for the detection and recognition of hazardous objects, including mines and explosive ammunition, based on combined terrain scanning with additional mathematical framework support with implementation of artificial intelligence methods. The proposed technology incorporates specialized algorithmic and hardware components and relies on the analysis of data acquired from unmanned aerial vehicle (UAV) surveys. It further involves ground-based re-inspection and comprehensive hazard assessment of previously identified areas of interest containing suspected objects. Mobile ground robotic systems are employed to perform multi-modal terrain scanning, followed by the construction of an aggregated data matrix, its subsequent analysis, and an integral evaluation using hyperspectral data processing techniques based on two-dimensional Fourier series. To identify hazardous objects with regular geometric shapes, dedicated artificial neural networks are utilized. The results of physical terrain scanning are represented within the framework of fuzzy set theory. As an illustrative example, an algorithm describing the application of an Elman neural network for the identification of circular-shaped hazardous objects is presented.

References

L. Marques et al., “State of the Art Review on Mobile Robots and Manipulators for Humanitarian Demining,” in Proc. 10th IARP Workshop Humanitarian Demining (HUDEM), Sibenik, Croatia, 2012.
V. A. Kyrylenko and V. R. Neroba, “Global problem of demining: state and approaches to solution,” Zb. Nauk. Pr. Cent. Voy-enno-Stratehichnykh Doslidzh. NUOU, no. 2 (66), pp. 115–119, 2020. doi: https://doi.org/10.33099/2304-2745/2019-2-66/115-119.
M. O. Popov, “Technology of remote detection of mines based on the analysis of survey materials from unmanned aerial vehi-cles: state and prospects,” Visn. Nac. Akad. Nauk Ukr., no. 5, pp. 56–62, May 2022, doi: https://doi.org/10.15407/visn2022.05.056.
B. A. Swett, A. J. Llorens and E. N. Hahn, “Designing Robots for the Battlefield: State of the Art,” in Designing Robots for the Battlefield, Cham, Switzerland: Springer, 2021, pp. 131–146, doi: https://doi.org/10.1007/978-3-030-54173-6_11.
G. Li, Z. Song and Q. Fu, “A New Method of Image Detection for Small Datasets under the Framework of YOLO Network,” in Proc. 2018 IEEE 3rd Adv. Inf. Technol., Electron. and Autom. Control Conf. (IAEAC), Oct. 2018, pp. 651–654, doi: https://doi.org/10.1109/iaeac.2018.8577214.
T. V. Pisarenko et al., Analysis of World Technological Trends in the Military Sphere, Kyiv, Ukraine: UkrISTEI, 2021, doi: https://doi.org/10.35668/978-966-479-127-1.
G. Fedorenko, H. Fesenko and V. Kharchenko, “Analysis of methods and development of the concept of guaranteed detection and recognition of explosive objects,” Inov. Tehnol. ta Nauk. Rishennya dlya Promislovosti, no. 4 (22), pp. 20–31, Dec. 2022. doi: https://doi.org/10.30837/itssi.2022.22.020.
S. Molochko, V. Bashynskyi, O. Kalamurza and V. Zhurakhov, “Analysis of the current state, characteristics and prospects of development of explosive ordnance detection sensors mounted on unmanned aerial systems,” Sci. Pap. State Sci. Res. Inst. Test. Certif. Weapon. Mil. Equip., no. 8, pp. 80–90, Jun. 2021, doi: https://doi.org/10.37701/dndivsovt.8.2021.09.
V. B. Strutynskyi and A. M. Hurzhii, Nazemni robotyzovani kompleksy: Monohrafiia. Zhytomyr, Ukraine: Ruta, 2023.
Y. Guo, M. Liao, W. Gao and G. Nie, “2D Semantic Segmentation: Recent Developments and Future Directions,” Future Inter-net, vol. 15, no. 6, p. 205, Jun. 2023. doi: https://doi.org/10.3390/fi15060205.
Li, Y., Fan, Q., Huang, H., Han, Z., & Gu, Q. (2023). A Modified YOLOv8 Detection Network for UAV Aerial Image Recog-nition. Drones, 7(5), 304, doi: https://doi.org/10.3390/drones7050304.
Y. Shi et al., “FBI-Pose: Towards Bridging the Gap between 2D Images and 3D Human Poses using Forward-or-Backward Information,” arXiv.org, 2018. [Online]. Accessed: Dec. 04, 2025. Available: https://arxiv.org/abs/1806.09241.
O. Maksymovych, “Speech recognition from Google Cloud: why use this service,” Cloudfresh, Oct. 27, 2023. [Online]. Ac-cessed: Dec. 04, 2025. Available: https://cloudfresh.com/ua/cloud-blog/google-speech-to-text-navishcho-vykorystovuvaty/.
V. B. Strutynskyi, Iu. M. Danylchenko, Iu. Y. Besarabets, O. A. Plivak and V. S. Pavlun, “Special ground robotic complexes for the implementation of demining technology based on artificial intelligence,” in Proc. Conf. Mech. Process. Mach. Tools In-strum., Zhytomyr, Ukraine, 2023, pp. 90–96. [Online]. Available: https://conf.ztu.edu.ua/protsesy-mekhanichnoyi-obrobky-verstaty-ta-instrument/.
V. B. Strutynskyi, Iu. M. Danylchenko, A. M. Hurzhii, S. V. Maidaniuk and V. S. Pavlun, “Application of artificial intelligence for terrain demining by special ground robotic complexes,” in Proc. Conf. Mech. Process. Mach. Tools Instrum., Zhytomyr, Ukraine, 2023, pp. 96–103. [Online]. Available: https://conf.ztu.edu.ua/protsesy-mekhanichnoyi-obrobky-verstaty-ta-instrument/.
V. B. Strutynskyi, A. M. Hurzhii, S. Iu. Vakulenko, and V. V. Novak, “Mobile robotic complex with remote control,” UA Patent 127035, Mar. 22, 2023.
Z. Miljković, M. Mitić, M. Lazarević, and B. Babić, “Neural network Reinforcement Learning for visual control of robot manip-ulators,” Expert Syst. Appl., vol. 40, no. 5, pp. 1721–1736, Sep. 2012, doi: https://doi.org/10.1016/j.eswa.2012.09.010.