VISUAL NAVIGATION OF UNMANNED AERIAL VEHICLES USING SEMANTIC TERRAIN DESCRIPTIONS
Abstract
The article addresses the problem of visual navigation for unmanned aerial vehicles (UAVs), which involves the automatic determination of the current position of the UAV (coordinates in the ground (local) coordinate system) based on the comparison and identification of descriptions of the current images (CI) received on board with reference descriptions stored in the form of a digital map in the memory of the UAV's onboard computer. The aim of this work is to improve the efficiency of visual navigation methods in terms of increasing computational performance, robustness, and accuracy of image identification algorithms in complex and changing observation conditions by using semantic descriptions of observed scenes. In this work, semantic descriptions are understood as descriptions that include classes of objects observed in the scene, their attributes, and relationships between them. The preparation of semantic descriptions of the map is carried out at the pre-flight preparation stage of the UAV using pre-trained neural networks for semantic segmentation. Semantic descriptions of the received CIs are generated on board the UAV. The use of neural network algorithms allows this process to be implemented in real-time for a wide range of observation conditions (different times of day and year). The use of semantic descriptions of the map and CI reduces computations compared to traditional pixel-by-pixel matching of raster images. Semantic descriptions are compared by matching object classes, their attributes, and relationships. The work presents a general algorithm for visual navigation, the main stages of the methodology for forming semantic descriptions, and the algorithm for comparing and identifying semantic descriptions of CIs and map descriptions. A hierarchical algorithm for comparing and identifying images based on the sequential application of semantic and raster descriptions of observed scenes is proposed. It is shown that the use of the procedure for comparing semantic descriptions of CIs and maps by the classes of objects present significantly reduces the computations necessary for image identification
References
1. Веремеенко К.К., Желтов С.Ю., Ким Н.В. и др. Современные информационные технологии в
задачах навигации и наведения беспилотных маневренных летательных аппаратов / под ред.
М.Н. Красильщикова, Г.Г. Себрякова. – М.: Физматлит, 2009. – 556 с.
2. Али Б., Садеков Р.Н., Цодокова В.В. Алгоритмы навигации беспилотных летательных аппаратов
с использованием систем технического зрения // Гироскопия и навигация. – 2022. – Т. 30, № 4
(119). – С. 87-105.
3. Красовский А.А., Белоглазов И.Н, Чигин Г.П. Теория корреляционно-экстремальных навигаци-
онных систем. – М.: Наука, 1979.
4. Баклицкий В.К., Бочкарев А.М., Мусьяков М.П. Методы фильтрации сигналов в корреляционно
экстремальных системах навигации. – М.: Радио и связь, 1988.
5. Ким Н.В. Обработка и анализ изображений в системах технического зрения: учеб. пособие.
– М.: Изд-во МАИ, 2001. – 164 с.
6. Горелик А.Л., Скрипкин В.А. Методы распознавания. – М.: Высшая школа, 1989. – 232 с.
7. Команда Лаборатории приборов, 2021, Оценка ошибки вычисления навигации инерциальными
модулями. – https://mp-lab.ru/navigation-error-estimation/?ysclid=lyg5vzaxqg875561868 (21 August
2024).
8. Pazychev D., Sadekov R. Simulation of INS Errors of Various Accuracy Classes. – 2020. – 1-3.
– https://doi.org/10.23919/ICINS43215.2020.9133869.
9. Семантическая сегментация изображений с использованием видимого и ближнего инфракрас-
ного каналов. – URL: https://habr.com/ru/articles/723646/ (дата обращения: 17.10.2024).
10. Vijay Badrinarayanan, Alex Kendall. SegNet: A Deep Convolutional EncoderDecoder Architecture
for Image Segmentation. – 2016. – C. 14.
11. Liang-Chieh Chen, George Papandreou. DeepLab: Semantic Image Segmentation with Deep Convolutional
Nets, Atrous Convolution, and Fully Connected CRFs. – 2017. – C. 14.
12. Shotton J., Winn J., Rother C., and Criminisi A. Textonboost for image understanding: Multi-class
object recognition and segmentation by jointly modeling texture, layout, and context // IJCV. – 2009.
13. Shotton J., Johnson M., and Cipolla R. Semantic texton forests for image categorization and segmentation
// in CVPR. – 2008.
14. Fulkerson B., Vedaldi A., and Soatto S. Class segmentation and object localization with superpixel
neighborhoods // in ICCV. – 2009.
15. Старовойтов В.В., Голуб Ю.И. Об оценке результатов классификации несбалансированных
данных по матрице ошибок // Информатика. – 2021. – Т. 18, № 1. – C. 61-71.
16. Лукашик Д.В. Анализ современных методов сегментации изображений // Экономика и качество
систем связи. – 2022. – № 2.
17. Zhongzhi Shi. Intelligence Science. Leading the Age of Intelligence. – Tsinghua University Press,
2021. – https://doi.org/10.1016/C2020-0-02066-9.
18. Коновалов К.Д. Алгоритм планирования маршрута БПЛА в условиях недостатка визуальных
ориентиров // Системы анализа и обработки данных. – 2024. – № 2 (94). – С. 37-54.
19. Zheng G., Jiang Z., Zhang1 H., Yao X. Deep semantic segmentation of unmanned aerial vehicle remote
sensing images based on fully convolutional neural network // Frontiers in Earth Science. – 2023.
– 11. – P. 12.
20. Pratap Singh K., Salvador J., Weihs L., Kembhavi A. Scene Graph Contrastive Learning for Embodied
Navigation // IEEE/CVF International Conference on Computer Vision (ICCV). – 2023. – P. 10884-
10894.
