Multi-Input CNN for Vision-Based In Situ Analysis of Extraterrestrial Surface Composition

Downloads

Authors

  • Kacper MARCINIAK Faculty of Mechanical Engineering, Wrocław University of Science and Technology, Wrocław, Poland ORCID ID 0009-0000-7098-5907
  • Michał GRZESIAK Faculty of Geoengineering, Mining and Geology, Wrocław University of Science and Technology, Wrocław, Poland ORCID ID 0009-0009-7749-3222
  • Igor ZAWORSKI Student Association of Unconventional Vehicles OFF-ROAD, Wrocław University of Science and Technology, Wrocław, Poland
  • Dominik PAWLISZEWSKI Student Association of Unconventional Vehicles OFF-ROAD, Wrocław University of Science and Technology, Wrocław, Poland
  • Dominika ZYGARLICKA Student Association of Unconventional Vehicles OFF-ROAD, Wrocław University of Science and Technology, Wrocław, Poland
  • Michał WNUK Student Association of Unconventional Vehicles OFF-ROAD, Wrocław University of Science and Technology, Wrocław, Poland
  • Adrian ZAKRZEWSKI Faculty of Mechanical Engineering, Wrocław University of Science and Technology, Wrocław, Poland ORCID ID 0000-0001-9171-9915

Abstract

Earth’s natural resources are finite, which is why engineers and scientists are increasingly directing their attention to the extraction of materials from celestial bodies. Beyond the extraction itself, however, a major challenge remains the localization of valuable substances and the assessment of their quality in situ. In this article, we present a flexible and robust method for estimating the content of selected components in heterogeneous mixtures using RGB image processing. The proposed deep learning architecture achieves high prediction accuracy with root mean squared error (RMSE) of (0.190 ±0.024) %. The framework supports a variety of backbone architectures, including lightweight models, making it suitable for deployment on edge devices such as planetary rovers. Furthermore, the method is flexible, allowing for easy adaptation to other tasks, for example, the analysis of more complex mixtures or inference based on multi- or hyperspectral imagery.

Keywords:

machine vision, machine learning, edge computing, in situ resource utilization, planetary rover

References



  1. Henley S., Allington R., PERC, CRIRSCO, and UNFC: Minerals reporting standards and classifications, European Geologist, 36: 49–54, 2013.


  2. Sommariva A., Gori L., Chizzolini B., Pianorsi M., The economics of Moon mining, Acta Astronautica, 170: 712–718, 2020, https://doi.org/10.1016/j.actaastro.2020.01.042.


  3. Levinson A., Taylor R., Moon Rocks and Minerals, Pergamon Press, New York, 1989.


  4. McKay D., Williams R., A geologic assessment of potential lunar ores, [In:] Space Resources and Space Settlements, J. Billingham, W. Gilbreath, B. O’Leary [Eds.], NASA SP-428, pp. 243–255, NASA Ames Research Center, 1979.


  5. Hutson M., Notes of Lunar Ilmenite, Lunar and Planetary Laboratory, The University of Arizona, pp. A1–A6, 1989.


  6. Sverdrup H.U., Sverdrup A.E., An assessment of the global supply, recycling, stocks in use and market price for titanium using the WORLD7 model, Sustainable Horizons, 7: 100067, 2023, https://doi.org/10.1016/j.horiz.2023.100067.


  7. Carr B.B., Recovery of water or oxygen by reduction of lunar rock, AIAA Journal, 1(4): 921–924, 1963.


  8. Paige D.A. et al., The lunar reconnaissance orbiter diviner lunar radiometer experiment, Space Science Review, 150: 125–160, 2009, https://doi.org/10.1007/s11214-009-9529-2.


  9. Pieters C.M. et al., The Moon Mineralogy Mapper (M3) on Chandrayaan-1, Current Science, 96(4): 500–505, 2008.


  10. Surkov Y., Shkuratov Y., Kaydash V., Korokhin V., Videen G., Lunar ilmenite content as assessed by improved Chandrayaan-1 M3 data, Icarus, 341: 113661, 2020, https://doi.org/10.1016/j.icarus.2020.113661.


  11. Charette M.P., McCord T.B., Pieters C., Adams J.B., Application of remote spectral reflectance measurements to lunar geology classification and determination of titanium content of lunar soils, Solid Earth and Planets, 79(11): 1605–1613, 1974, https://doi.org/10.1029/JB079i011p01605.


  12. Lucey P.G., Blewett D.T., Jolliff B.L., Lunar iron and titanium abundance algorithms based on final processing Clementine ultraviolet-visible images, Journal of Geophysical Research, 105: 20297–20305, 2000, https://doi.org/10.1029/1999JE001117.


  13. Palafox L.F., Hamilton C.W., Scheidt S.P., Alvarez A.M., Automated detection of geological landforms on Mars using Convolutional Neural Networks, Computers & Geosciences, 101: 48–56, 2017, https://doi.org/10.1016/j.cageo.2016.12.015.


  14. Li J. et al., Autonomous Martian rock image classification based on transfer deep learning methods, Earth Science Informatics, 13: 951–963, 2020, https://doi.org/10.1007/s12145-019-00433-9.


  15. Zhou L., Liu Z., Wang W., Terrain classification algorithm for Lunar rover using a deep ensemble network with high-resolution features and interdependencies between channels, Wireless Communications and Mobile Computing, 2020: 8842227, 2020, https://doi.org/10.1155/2020/8842227.


  16. Lv F. et al., Highly accurate visual method of Mars terrain classification for rovers based on novel image features, Entropy, 24(9): 1304, 2022, https://doi.org/10.3390/e24091304.


  17. Razzell Hollis J. et al., The power of paired proximity science observations: Co-located data from SHERLOC and PIXL on Mars, Icarus, 387: 115179, 2022, https://doi.org/10.1016/j.icarus.2022.115179.


  18. Edgett K.S. et al., Curiosity’s robotic arm-mounted Mars Hand Lens Imager (MAHLI): Characterization and calibration status, MSL MAHLI Technical Report 0001, Mars Science Laboratory, 2015, https://doi.org/10.13140/RG.2.1.3798.5447.


  19. Herkenhoff K.E. et al., Evidence from opportunity’s microscopic imager for Water on Meridiani Planum, Science, 306(5702): 1727–1730, 2004, https://doi.org/10.1126/science.1105286.


  20. Xu T.Y., Hapke B., Zhang X.P., Wu Y.Z., Lu X.P., Micro-scale photometry of the Moon using Chang’E-3 Panoramic Camera (PCAM), Astronomy & Astrophysics, 665: A15, 2022, https://doi.org/10.1051/0004-6361/202143012.


  21. Zhong J., Yan J., Li M., Barriot J.-P., A deep learning-based local feature extraction method for improved image matching and surface reconstruction from Yutu-2 PCAM images on the Moon, ISPRS Journal of Photogrammetry and Remote Sensing, 206: 16–29, 2023, https://doi.org/10.1016/j.isprsjprs.2023.10.021.


  22. Els S. et al., The microscope camera CAM-M on-board the Rashid-1 Lunar rover, Space Science Reviews, 220: 81, 2024, https://doi.org/10.1007/s11214-024-01117-7.


  23. Oberbeck V.R., Hoerz F., Quaide W., Morrison R.H., Emplacement of the Caley Formation, Technical Report, NASA, 1973.


  24. Head J.W. et al., Lunar mare basaltic Volcanism: Volcanic features and emplacement processes, Reviews in Mineralogy and Geochemistry, 89(1): 453–507, 2023, https://doi.org/10.2138/rmg.2023.89.11.


  25. Steffens C.R., Messias L.R.V., Drews P.L.J., Botelho S.S.d.C., Can exposure, noise and compression affect image recognition? An assessment of the impacts on state-of-the-art ConvNets, [In:] 2019 Latin American Robotics Symposium (LARS), 2019 Brazilian Symposium on Robotics (SBR) and 2019 Workshop on Robotics in Education (WRE), Rio Grande, Brazil, pp. 61–66, 2019, https://doi.org/10.1109/LARS-SBR-WRE48964.2019.00019.


  26. Seals M., On the Robustness of Object Detection Based Deep Learning Models, MA thesis, University of Tennessee, 2019.


  27. Dai W., Berleant D., Benchmarking robustness of deep learning classifiers using two-factor perturbation, [In:] 2021 IEEE International Conference on Big Data (Big Data), Orlando, FL, USA, pp. 5085–5094, 2021, https://doi.org/10.1109/BigData52589.2021.9671976.