V.A. Pavlov1, A.K. Tsytsulin2, A.I. Bobrovsky3, Yu.I. Mukalo4, D.N. Bechin5, Ya.Yu. Belogubkin6

1 Peter the Great St. Petersburg Polytechnic University (St. Petersburg, Russia)

2,5,6 Television Research Institute (St. Petersburg, Russia)

3,4 Government Scientific Research Institute of Applied Problems (St. Petersburg, Russia)

1 pavlov_va@spbstu.ru; 2 atsytsulin@mail.ru; 3 albob@mail.ru; 4 metoi@yandex.ru; 5 d.bechin@niitv.ru; 6 j.belogubkin@niitv.ru

Currently, the transition to autonomous intelligent spacecraft control systems requires the development of reliable computer vision algorithms for the tasks of rendezvous, docking and monitoring of objects in near-Earth space. However, the creation of effective neural network models faces the problem of a shortage of full-scale spacecraft images obtained directly in space. This paper explores the possibility of overcoming this deficit through the use of a synthetic approach based on three-dimensional models. The data generation process included processing nine spacecraft models of various classes. The synthesis technique involved rendering models with a rotation step of 30° along three axes, random scaling in the range from 50% to 150%, and superimposing on a variable background simulating space. The use of monochrome images made it possible to minimize excessive color information and focus training on the geometric features of the devices. The modern YOLOv11 architecture was chosen to detect the spacecraft. The experimental assessment on the test sample demonstrated an average accuracy of mAP50 at the level of 0.994 and mAP50-95 at the level of 0.986. At the same time, for devices with pronounced geometry, such as AcrimSAT and AIM, the mAP50 values reached 0.995.

Pavlov V.A., Tsytsulin A.K., Bobrovsky A.I., Mukalo Yu.I., Bechin D.N., Belogubkin Ya.Yu. Automatic detection of spacecraft using synthetic data. Radiotekhnika. 2026. V. 90. № 3. P. 107−116. DOI: https://doi.org/10.18127/j00338486-202603-09 (In Russian)

1. Pavanello Z., De Maria L., De Vittori A., Maestrini M., Di Lizia P., Armellin R. CAMmary: a review of spacecraft collision avoidance manoeuvre design methods. Acta Astronautica. 2025. V. 236. P. 770–789.
2. Zhukov A.O., Gedzon V.S., Pereverzev A.F., Bashkatov A.I. Analiz funkcional'nyh vozmozhnostej sistemy nablyudeniya kosmicheskih ob"ektov v slozhnoj fono-celevoj obstanovke. Okolozemnaya astronomiya. 2022. S. 53-55 (in Russian).
3. Kisantal M., Sharma S., Park T. H., Izzo D., Märtens M., D'Amico S. Satellite pose estimation challenge: dataset, competition design and results. arXiv preprint arXiv:1911.02050. 2020.
4. Park T.H., Märtens M., Lecuyer G., Izzo D., D'Amico S. SPEED+: next-generation dataset for spacecraft pose estimation across domain gap. 2022 IEEE Aerospace Conference (AERO). Big Sky. MT. USA. 2022. Р. 1‒15.
5. Proenca P.F., Gao Y. Deep learning for spacecraft pose estimation from photorealistic rendering. arXiv preprint arXiv:1907.04298. 2019.
6. AlDahoul N., Karim, H.A., Momo M.A. RGB-D based multi-modal deep learning for spacecraft and debris recognition. Scientific Reports. 2022. V. 12. № 1. P. 3924.
7. Hematulin W., Kamsing P., Phisannupawong T., Panyalert T., Manuthasna S., Torteeka P., Boonsrimuang P. Generating large-scale datasets for spacecraft pose estimation via a high-resolution synthetic image renderer. Aerospace. 2025. V. 12. № 4. P. 334.
8. 3D resources - NASA science [электронный ресурс]. NASA Science. 2025. URL: https://science.nasa.gov/3d-resources/ (дата обращения: 30.12.2025).
9. Amjoud A.B., Amrouch M. Object detection using deep learning, CNNs and vision transformers: a review. IEEE Access. 2023. V. 111. P. 35479–35516.
10. Khanam R., Hussain M. YOLOv11: an overview of the key architectural enhancements. arXiv preprint, arXiv:2410.17725v1. 2024.
11. Everingham M., Van Gool L., Williams C.K.I., et al. The pascal visual object classes (VOC) challenge. International Journal of Computer Vision. 2010. V. 88. № 2. P. 303–338.
12. Lin T.-Y., Maire M., Belongie S., Hays J., Perona P., Ramanan D., Dollár P., Zitnick C.L. Microsoft COCO: common objects in context. Computer Vision – ECCV 2014: Proceedings of the 13th European Conference on Computer Vision. Cham: Springer International Publishing. 2014. P. 740–755.
13. Shustov B.M., Zolotaryov R.V., Shugarov A.S. Ob astronomicheskih obosnovaniyah tekhnicheskih sredstv obnaruzheniya asteroidov, sblizhayushchihsya s Zemlej. Nauchnye trudy INASAN. 2025. T. 10. № 3. S. 122–134 (in Russian).