<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">inttrans</journal-id><journal-title-group><journal-title xml:lang="ru">Интеллектуальный транспорт</journal-title><trans-title-group xml:lang="en"><trans-title>Intelligent transport</trans-title></trans-title-group></journal-title-group><issn pub-type="epub">3033-6007</issn><publisher><publisher-name>АО «НИИАС»</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.24412/3033-6007-2026-339-23-36</article-id><article-id custom-type="elpub" pub-id-type="custom">inttrans-114</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>ИСКУССТВЕННЫЙ ИНТЕЛЛЕКТ И МАШИННОЕ ОБУЧЕНИЕ</subject></subj-group></article-categories><title-group><article-title>Нейросетевые методы распознавания подземных объектов в данных георадара мобильного колесного робота</article-title><trans-title-group xml:lang="en"><trans-title>Neural Network Methods for Recognizing Underground Objects in the Data of a Mobile Wheeled Robot's Ground-Penetrating Radar</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Кривошей</surname><given-names>М. И.</given-names></name><name name-style="western" xml:lang="en"><surname>Krivoshey</surname><given-names>M. I.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Техник</p></bio><bio xml:lang="en"><p>Technician</p></bio><email xlink:type="simple">krivoshei.mi@mipt.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Юдин</surname><given-names>Д. А.</given-names></name><name name-style="western" xml:lang="en"><surname>Yudin</surname><given-names>D. A.</given-names></name></name-alternatives><bio xml:lang="ru"><p>к.т.н., заведующий лабораторией интеллектуального транспорта МФТИ-НКБВС</p></bio><bio xml:lang="en"><p>Ph.D., Head of the Laboratory of Intelligent Transport at MIPT‑NKBVS</p></bio><email xlink:type="simple">yudin.da@mipt.ru</email><xref ref-type="aff" rid="aff-2"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Зарипов</surname><given-names>М. Н.</given-names></name><name name-style="western" xml:lang="en"><surname>Zaripov</surname><given-names>M. N.</given-names></name></name-alternatives><bio xml:lang="ru"><p>ведущий программист‑разработчик</p></bio><bio xml:lang="en"><p>lead software developer</p></bio><email xlink:type="simple">zaripov.mn@mipt.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Буличев</surname><given-names>О. В.</given-names></name><name name-style="western" xml:lang="en"><surname>Bulichev</surname><given-names>O. V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>к.т.н.</p></bio><bio xml:lang="en"><p>Ph.D.</p></bio><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Муравьев</surname><given-names>К. Ф.</given-names></name><name name-style="western" xml:lang="en"><surname>Muravyev</surname><given-names>K. F.</given-names></name></name-alternatives><bio xml:lang="ru"><p>к.ф.-м.н., научный соmудник</p></bio><bio xml:lang="en"><p>Candidate of Physical and Mathematical Sciences, Researcher</p></bio><email xlink:type="simple">kirill.mouraviev@yandex.ru</email><xref ref-type="aff" rid="aff-3"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Панов</surname><given-names>А. И.</given-names></name><name name-style="western" xml:lang="en"><surname>Panov</surname><given-names>A. I.</given-names></name></name-alternatives><bio xml:lang="ru"><p>д.ф.-м.н.</p></bio><bio xml:lang="en"><p>Doctor of Physical and Mathematical Sciences</p></bio><email xlink:type="simple">panov.ai@mipt.ru</email><xref ref-type="aff" rid="aff-2"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Московский физико-технический институт (МФТИ)</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Moscow Institute of Physics and Technology (MIPT)</institution><country>Russian Federation</country></aff></aff-alternatives><aff-alternatives id="aff-2"><aff xml:lang="ru"><institution>Московский физико-технический институт (МФТИ); Институт искусственного интеллекта AIRI</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Moscow Institute of Physics and Technology (MIPT); Artificial Intelligence Research Institute (AIRI)</institution><country>Russian Federation</country></aff></aff-alternatives><aff-alternatives id="aff-3"><aff xml:lang="ru"><institution>Федеральный исследовательский центр «Информатика и управление» РАН (ФИЦ ИУ РАН); Московский физико-технический институт (МФТИ)</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Federal Research Center «Computer Science and Control» of the RAS (FRC CSC RAS); Moscow Institute of Physics and Technology (MIPT)</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2026</year></pub-date><pub-date pub-type="epub"><day>27</day><month>09</month><year>2026</year></pub-date><volume>10</volume><issue>3(39)</issue><fpage>23</fpage><lpage>36</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Кривошей М.И., Юдин Д.А., Зарипов М.Н., Буличев О.В., Муравьев К.Ф., Панов А.И., 2026</copyright-statement><copyright-year>2026</copyright-year><copyright-holder xml:lang="ru">Кривошей М.И., Юдин Д.А., Зарипов М.Н., Буличев О.В., Муравьев К.Ф., Панов А.И.</copyright-holder><copyright-holder xml:lang="en">Krivoshey M.I., Yudin D.A., Zaripov M.N., Bulichev O.V., Muravyev K.F., Panov A.I.</copyright-holder><license xml:lang="ru" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>Данная работа распространяется под лицензией Creative Commons Attribution 4.0.</license-p></license><license xml:lang="en" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>This work is licensed under a Creative Commons Attribution 4.0 License.</license-p></license></permissions><self-uri xlink:href="https://www.intelligent-transport.ru/jour/article/view/114">https://www.intelligent-transport.ru/jour/article/view/114</self-uri><abstract><p>В настоящее время для решения задач, связанных с поиском и распознаванием объектов под землёй, часто используется георадар. Это позволяет получать богатую информацию об исследуемой местности на различной глубине, однако ручная интерпретация радарограмм связана с большими трудозатратами. В данной работе предлагается метод GeoRTr для автоматического нахождения подземных объектов интереса на основе нейронной сети оригинальной архитектуры с механизмом внимания GeoAttention, на вход которой подаются A-сканы георадара, установленного на колёсном роботе, перемещающемся в автоматическом либо полуавтоматическом режиме. Предлагаемая нейросетевая модель позволяет учитывать пространственные взаимосвязи между A-сканами георадара даже в тех случаях, когда траектория движения робота отличается от прямой линии. С использованием мобильного колёсного робота и георадара, антенна которого установлена на небольшой высоте над анализируемой поверхностью, собран и размечен уникальный набор данных для распознавания объектов нескольких категорий, находящихся под поверхностью земли. Эксперименты с разработанной нейросетевой моделью на собранном наборе данных продемонстрировали F1-меру классификации свыше 40%, что превосходит по качеству методы-аналоги. В ходе сравнительного анализа различных вариаций архитектуры модели выделены две конфигурации для разных условий решения задачи георадарного исследования.</p></abstract><trans-abstract xml:lang="en"><p>Currently, ground-penetrating radar (GPR) is often used to solve problems related to searching for and recognizing underground objects. This allows for obtaining rich information about the studied area at various depths, but manual interpretation of radargrams is labor-intensive. This paper proposes the GeoRTr method for automatically detecting underground objects of interest. It is based on an original neural network architecture with a GeoAttention mechanism. The network's input is A-scans from a GPR mounted on a wheeled robot moving in automatic or semi-automatic mode. The proposed neural network model accounts for spatial relationships between A-scans even when the robot's trajectory deviates from a straight line. Using a mobile wheeled robot and a GPR with an antenna mounted at a low altitude above the surface being analyzed, a unique dataset was collected and labeled for recognizing several categories of subsurface objects. Experiments with the developed neural network model on the collected dataset demonstrated a classification F1-score of over 40%, surpassing comparable methods. A comparative analysis of various model architecture variations identified two configurations for different ground-penetrating radar object search conditions.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>мобильный колёсный робот</kwd><kwd>A-сканы</kwd><kwd>георадар</kwd><kwd>нейронная сеть</kwd><kwd>трансформер</kwd><kwd>распознавание подземных объектов</kwd></kwd-group><kwd-group xml:lang="en"><kwd>mobile wheeled robot</kwd><kwd>A-scans</kwd><kwd>ground-penetrating radar</kwd><kwd>neural network</kwd><kwd>transformer</kwd><kwd>underground object recognition</kwd></kwd-group><funding-group><funding-statement xml:lang="ru">Работа частично выполнена в рамках НИР «Разработка метода одновременной локализации и картирования TerraCognitaSLAM на основе георадара и визуально-инерциальных сенсоров для условий недоступного спутникового сигнала» (по Договору между Центральным университетом и МФТИ).   Авторы выражают благодарность Центру робототехники Сбера за предоставление робота AgileX Scout 2.0 для проведения экспериментов, а также Шишкову Д. Л. за предоставление георадарного оборудования для сбора набора данных, описанного в настоящей статье.</funding-statement><funding-statement xml:lang="en">The work was partially carried out as part of the research project «Development of the TerraCognitaSLAM simultaneous localization and mapping method based on ground‑penetrating radar and visual‑inertial sensors for conditions with an inaccessible satellite signal» (under the Agreement between the Central University and MIPT).  The authors express their gratitude to the Sber Robotics Center for providing the AgileX Scout 2.0 robot for conducting the experiments, as well as to D. L. Shishkov for providing the ground‑penetrating radar equipment for collecting the dataset described in this article.</funding-statement></funding-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Wu, M. Two-stage GPR image inversion method based on multi-scale dilated convolution and hybrid attention gate / M. Wu, Q. Liu, S. Ouyang // Remote Sensing. – 2025. – Vol. 17. – No. 2. – Art. 322. – DOI 10.3390/rs17020322.</mixed-citation><mixed-citation xml:lang="en">Wu, M., Liu, Q., &amp; Ouyang, S. (2025). Two-stage GPR image inversion method based on multi-scale dilated convolution and hybrid attention gate. Remote Sensing, 17(2), Article 322. https://doi.org/10.3390/rs17020322</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Lee, K. GPR-Former: Context-Aware Moisture Detection / K. Lee, C. Feng // ISARC. Proceedings of the International Symposium on Automation and Robotics in Construction. – Montreal: IAARC Publications, 2025. – Vol. 42. – P. 1–8. – DOI 10.22260/isarc2025/0002.</mixed-citation><mixed-citation xml:lang="en">Lee, K., &amp; Feng, C. (2025). GPR-Former: Context-aware moisture detection. In ISARC. Proceedings of the International Symposium on Automation and Robotics in Construction (pp. 1–8). IAARC Publications. https://doi.org/10.22260/isarc2025/0002</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Jeon, M. Efficient and Robust Temporal Modeling for LGPR Odometry With Mamba / M. Jeon, H. Lee, J. Lee // IEEE Access. – 2026. – Vol. 14. – P. 38775–38792. – DOI 10.1109/ACCESS.2026.3672677.</mixed-citation><mixed-citation xml:lang="en">Jeon, M., Lee, H., &amp; Lee, J. (2026). Efficient and robust temporal modeling for LGPR odometry with Mamba. IEEE Access, 14, 38775–38792. https://doi.org/10.1109/ACCESS.2026.36</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Automated railway subgrade defect detection using GPR and an enhanced YOLOv11 framework / M. T. Tegegne, R. Yang, S. Huang [et al.] // Transportation Geotechnics. – 2026. – Vol. 62. – Art. 102105. – DOI 10.1016/j.trgeo.2026.102105.</mixed-citation><mixed-citation xml:lang="en">Tegegne, M. T., Yang, R., Huang, S., Xiao, H., Yang, Z., Zhang, Q., Liu, Y., &amp; Qi, W. (2026). Automated railway subgrade defect detection using GPR and an enhanced YOLOv11 framework. Transportation Geotechnics, 62, Article 102105. https://doi.org/10.1016/j.trgeo.2026.1</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Lei, W. HFL-YOLOv8: a hyperbolic feature-enhanced lightweight network for object detection in ground penetrating radar images / W. Lei, S. Yu, T. Zhang // Applied Soft Computing. – 2026. – Vol. 188. – Art. 114403. – DOI 10.1016/j.asoc.2025.114403.</mixed-citation><mixed-citation xml:lang="en">Lei, W., Yu, S., &amp; Zhang, T. (2026). HFL-YOLOv8: A hyperbolic feature-enhanced lightweight network for object detection in ground penetrating radar images. Applied Soft Computing, 188, Article 114403. https://doi.org/10.1016/j.asoc.2025.114403</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">TLAD-YOLO: Lightweight network for intelligent detection of railway tunnel lining anomalies using ground penetrating radar / P. Zheng, A. Zhang, Z. Shi [et al.] // Journal of Applied Geophysics. – 2025. – Vol. 241. – Art. 105869. – DOI 10.1016/j.jappgeo.2025.105869.</mixed-citation><mixed-citation xml:lang="en">Zheng, P., Zhang, A., Shi, Z., Wang, S., Ma, Y., &amp; Liu, Z. (2025). TLAD-YOLO: Lightweight network for intelligent detection of railway tunnel lining anomalies using ground penetrating radar. Journal of Applied Geophysics, 241, Article 105869. https://doi.org/10.1016/j.jappgeo.2025.105869</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">ME-YOLO: A novel real-time detection network for pavement interlayer distress using ground penetrating radar / S. Cao, C. Lu, X. Wang [et al.] // Journal of Applied Geophysics. – 2026. – Vol. 245. – Art. 106057. – DOI 10.1016/j.jappgeo.2025.106057.</mixed-citation><mixed-citation xml:lang="en">Cao, S., Lu, C., Wang, X., Zhang, P., Jin, G., &amp; Cai, W. (2026). ME-YOLO: A novel real-time detection network for pavement interlayer distress using ground-penetrating radar. Journal of Applied Geophysics, 245, Article 106057. https://doi.org/10.1016/j.jappgeo.2025.106</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">FLS-YOLO: a multi-scale subsurface defect detection network via heterogeneous grouping and spatial reconstruction / W. Zhang, Z. Liu, Q. Yang, X. Qiu // Signal, Image and Video Processing. – 2026. – Vol. 20. – No. 5. – Art. 262. – DOI 10.1007/s11760-026-05352-z.</mixed-citation><mixed-citation xml:lang="en">Zhang, W., Liu, Z., Yang, Q., &amp; Qiu, X. (2026). FLS-YOLO: A multi-scale subsurface defect detection network via heterogeneous grouping and spatial reconstruction. Signal, Image and Video Processing, 20(5), Article 262. https://doi.org/10.1007/s11760-026-05352-z</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">A Novel Shape-Aware Topological Representation for GPR Data with DNN Integration / M. Kang, S. Kaji, S.-Y. Lee [et al.] // arXiv. – 2025. – arXiv:2506.06311. – DOI 10.48550/arXiv.2506.06311.</mixed-citation><mixed-citation xml:lang="en">Kang, M., Kaji, S., Lee, S.-Y., Kim, T., Ryu, H.-H., &amp; Choi, S. (2025). A novel shapeaware topological representation for GPR data with DNN integration (arXiv:2506.06311). arXiv. https://doi.org/10.48550/arXiv.2506.06311</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Arvas, M. M. Detection of underground objects from GPR data using a lightweight YOLO-based approach / M. M. Arvas, A. C¸ ınar // Scientific Reports. – 2026. – DOI 10.1038/s41598-026-59135-0.</mixed-citation><mixed-citation xml:lang="en">Arvas, M. M., &amp; Çınar, A. (2026). Detection of underground objects from GPR data using a lightweight YOLO-based approach. Scientific Reports. https://doi.org/10.1038/s41598-0-59135-0</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Intelligent recognition of GPR road hidden defect images based on feature fusion and attention mechanism / H. Lv, Y. Zhang, J. Dai [et al.] // IEEE Transactions on Geoscience and Remote Sensing. – 2025. – Vol. 63. – P. 1–17.</mixed-citation><mixed-citation xml:lang="en">Lv, H., Zhang, Y., Dai, J., Wu, H., Wang, J., &amp; Wang, D. (2025). Intelligent recognition of GPR road hidden defect images based on feature fusion and attention mechanism. IEEE Transactions on Geoscience and Remote Sensing, 63, 1–17.</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Lightweight CNN model for automatic detection and depth estimation of subsurface voids using GPR B-scan data / A. Mojahid, D. El Ouai, K. El Amraoui [et al.] // Natural Hazards Research. – 2025. – Vol. 5. – No. 2. – P. 432–446. – DOI 10.1016/j.nhres.2025.02.001.</mixed-citation><mixed-citation xml:lang="en">Mojahid, A., El Ouai, D., El Amraoui, K., El-Hami, K., Aitbenamer, H., Verrelst, J., &amp; Barone, P. M. (2025). Lightweight CNN model for automatic detection and depth estimation of subsurface voids using GPR B-scan data. Natural Hazards Research, 5(2), 432–446. https://doi.org/10.1016/j.nhres.2025.02.001</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">GPR-TSBiNet: An information gradient enrichment model for GPR B-scan small target detection / C. Wang, Y. Guan, M. Chi [et al.] // Sensors. – 2025. – Vol. 25. – No. 7. – Art. 2223. – DOI 10.3390/s25072223.</mixed-citation><mixed-citation xml:lang="en">Wang, C., Guan, Y., Chi, M., Shen, F., Yu, Z., Chen, Q., &amp; Chen, C. (2025). GPR-TSBiNet: An information gradient enrichment model for GPR B-scan small target detection. Sensors, (7), Article 2223. https://doi.org/10.3390/s25072223</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Attention-Module-Guided Time-Lapse Leakage Plume Imaging Driven by LeakInv-CUNet GPR Inversion Framework / H. Wang [et al.] // IEEE Access. – 2025. – Vol. 13. – P. 122514–122529.</mixed-citation><mixed-citation xml:lang="en">Wang, H., et al. (2025). Attention-module-guided time-lapse leakage plume imaging driven by LeakInv-CUNet GPR inversion framework. IEEE Access, 13, 122514–122529.</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Enhanced Pix2pixGAN with Spatial-Channel Attention for Underground Medium Inversion from GPR / S. Yang, L. Guo, Y. Yang, H. Ye // Remote Sensing. – 2026. – Vol. 18. – No. 3. – Art. 448. – DOI 10.3390/rs18030448.</mixed-citation><mixed-citation xml:lang="en">Yang, S., Guo, L., Yang, Y., &amp; Ye, H. (2026). Enhanced Pix2pixGAN with spatial-channel attention for underground medium inversion from GPR. Remote Sensing, 18(3), Article 448. https://doi.org/10.3390/rs18030448</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
