AI for GNSS Remote Sensing

Joint Study Group 1

Joint with IAG Sub-Commission 4.3, GGOS FA GSWR, and ICCC

Chair: Milad Asgarimehr (Germany),

Vice-Chair: Lei Liu (USA),

General Contact:

Terms of Reference (ToR) / Description: 

GNSS remote sensing exploits the signals transmitted by positioning and navigation systems. GNSS provides visibility to multiple GNSS satellites at any point on the globe, enabling Earth observations with high spatial coverage and sampling frequency. GNSS remote sensing techniques, such as GNSS Reflectometry (GNSS-R), Radio Occultation (GNSS-RO), and ground-based sounding approaches, require only a receiver component as they exploit GNSS signals of opportunity. Consequently, the use of GNSS receivers for remote sensing purposes has rapidly grown in regional and global ground networks, as well as in small satellites and CubeSat constellations. These receivers generate large datasets that offer significant potential for data-driven algorithm development and “learning” by AI models.

However, the full potential of AI in GNSS remote sensing has not yet been realised. In contrast, machine learning techniques are well established in other remote sensing domains, particularly in the processing of optical Earth observations. AI can enhance retrieval and modelling capabilities, which is particularly valuable in GNSS remote sensing, especially where theoretical models lack validation under different field conditions, or when the associated physics cannot be easily formulated using theoretical knowledge. This study group aims to use AI approaches to drive innovative and interdisciplinary developments in GNSS Earth observations and Earth system modelling, including land/ocean surface parameters, atmospheric, ionospheric, and space weather processes.


  • Explore the potential of advanced AI techniques for retrieving Earth’s surface, atmospheric, climate, ionospheric, and space weather parameters from GNSS remote sensing data as a sole source of information or in fusion with data from other sensors/techniques. The studies may include quantification of model uncertainty.
  • Study physics-informed AI to embed theoretical knowledge in the learning process and to mutually improve physical understanding and domain knowledge through explainable modelling, inverse knowledge extraction, pattern recognition, and simulation.
  • Explore the potential for prediction of surface, atmospheric, and ionospheric patterns and climate-driven changes by training AI on historical data and incorporating real-time information.
  • Leverage AI models for Earth system modelling and forecasting using GNSS data to support risk assessment, early warning systems, and decision making for climate change mitigation and adaptation.


  1. Fikri Bamahry (Reference systems and Planetology Department, Royal Observatory of Belgium, Belgium,
  2. Mohammad Sharifi (School of Surveying and Geospatial Engineering, University of Tehran, Iran,
  3. Grzegorz Nykiel (Faculty of Civil and Environmental Engineering, Gdansk University of Technology, Poland,
  4. Randa Natras (Department of Aerospace and Geodesy, Technical University of Munich, Germany,
  5. Manuel Martin-Neira (European Space Agency, the Netherlands,
  6. Zorheh Adavai (Department of Geodesy and Geoinformation, TU Wien,
  7. Artem Smirnov (German Research Centre for Geosciences GFZ, Germany,
  8. Jonathan Jones (Met Office, UK,
  9. Daixin Zhao (Technical University of Munich & German Aerospace Center DLR, Germany,
  10. Shuyin Mao (Institute of Geodesy and Photogrammetry, ETH Zurich,
  11. Möller Gregor (Department of Geodesy and Geoinformation, TU Wien,
  12. Timothy Kodikara (German Aerospace Center DLR, Germany,
  13. Stylianos Kossieris (Institute of Communication and Computer Systems in Athens, Greece,
  14. Jens Wickert (German Research Centre for Geosciences GFZ, Germany,
  15. Caroline Arnold (German Climate Computing Centre DKRZ, Germany,
  16. Matthias Aichinger-Rosenberger (Institute of Geodesy and Photogrammetry, ETH Zurich,
  17. Tianqi Xiao (Institute of Geodesy and Geoinformation Science, TU-Berlin, Germany,
  18. Mario Moreno (German Aerospace Center DLR, Germany,
  19. Laura Crocetti (Institute of Geodesy and Photogrammetry, ETH Zurich,
  20. Cansu Beşel (Department of Geomatics Engineering, Sinop University, Turkey,
  21. Ole Roggenbuck (Federal Agency for Cartography and Geodesy BKG, Germany,
  22. Cheng Wang (Beihang University, China,
  23. Witold Rohm (Institute of Geodesy and Geoinformatics, Wroclaw University of Environmental and Life Sciences, Poland,
  24. Jihye Park (Oregon State University, USA,