You are here: / / / / / AI for Earth Orientation Parameter Prediction

AI for Earth Orientation Parameter Prediction

Joint Study Group 3

Joint with IAG Commission 3

Chair: Dr. Sadegh Modiri (BKG, Germany),

Vice-chair: Dr. Justyna Śliwińska-Bronowicz  (Polish Academy of Sciences, Poland),

Vice-chair: Dr. Santiago Belda (University of Alicante, Spain), " rel="nofollow" data-wpel-link="internal">

General Contact: ai4eop (at) ggos.org

Abstract

The third study group will explore the use of AI for predicting Earth orientation parameters. This group will build on the successful Second Earth Orientation Parameter Prediction Comparison Campaign organized by the International Earth Rotation and Reference Systems Service (IERS) and will continue to investigate machine learning for the prediction of Earth orientation parameters and effective angular momentum.

Description – Terms of Reference (ToR)

The Artificial Intelligence for Earth Orientation Parameter Prediction study group recognizes the pivotal significance of accurate Earth orientation parameters (EOP) prediction and the transformative potential of artificial intelligence (AI) in advancing our understanding of
these parameters. EOP, encompassing nutation offsets, pole coordinates, and dUT1, serve as a crucial link between the terrestrial and celestial reference frames, playing a vital role in many operational activities. Navigation of deep-space satellite missions, the precise pointing of astronomical instruments, and satellite-based positioning on Earth heavily rely on precise EOP predictions. Timely and reliable estimates of these parameters are essential for ensuring the success and accuracy of such endeavours. Recognizing this need, numerous agencies and institutions worldwide diligently process space geodetic observations to obtain rapid estimates, forming the foundation for subsequent EOP predictions. The advent of AI presents an unparalleled opportunity to revolutionize EOP prediction by leveraging the remarkable progress made in AI methodologies. AI techniques, including natural language processing, computer vision, and deep learning, have demonstrated their prowess in various fields, enabling breakthroughs in data analysis, pattern recognition, and hypothesis generation. By harnessing these AI capabilities, scientists can delve deeper into complex EOP theories, enhance our understanding of Earth’s orientation, and refine prediction models. AI empowers researchers to analyze vast and intricate datasets with remarkable efficiency, unravelling hidden patterns and relationships that might have eluded traditional methods. These AI-driven insights accelerate scientific discovery and contribute to the development of robust EOP prediction models. The ability to process and interpret complex data, coupled with AI’s capacity for adaptive learning, opens new avenues for refining existing theories and exploring novel hypotheses related to Earth’s orientation. Through the collaborative efforts of the study group, AI methodologies are deployed to enhance the accuracy, reliability, and timeliness of EOP predictions. By integrating AI into the existing prediction processes, scientists can harness the power of machine learning algorithms to improve prediction models based on observed EOP data continually. This iterative feedback loop strengthens the foundations of EOP theories, enabling a more comprehensive understanding of Earth’s orientation dynamics. The study group’s primary objective is to foster innovation, collaboration, and knowledge exchange among experts in AI and geodesy. By bringing together interdisciplinary expertise, the group aims to push the boundaries of AI research, uncovering novel insights into EOP theories and further refining prediction methodologies. This collective effort enhances our understanding of Earth’s orientation and advances the accuracy of navigation, positioning, and astronomical endeavours that rely heavily on EOP predictions. In summary, the study group Artificial Intelligence for Earth Orientation Parameter Prediction recognizes the critical importance of accurate EOP predictions and embraces the transformative potential of AI. By synergistically merging AI methodologies with the rich field of geodesy, the group aims to unlock new frontiers in EOP research, thereby advancing our understanding of Earth’s orientation and improving the precision and reliability of EOP predictions. Through collaboration, exploration, and innovation, the study group strives to shape the future of EOP prediction and its applications in various operational activities, paving the way for significant advancements in navigation, positioning, and astronomical sciences.

Join us

Are you interested in participating in this Joint Study Group? Please simply fill out this form:


Objectives

Objectives of the study group ”AI for EOP Prediction”:

  • Investigate the results of the 2nd Earth Orientation Parameter Prediction Comparison Campaign (2nd EOP PCC): The study group aims to thoroughly investigate the results of 2nd EOP PCC, analyzing the accuracy, consistency, and reliability of the obtained parameters. By evaluating the performance of the EOP PCC method, the group aims to identify areas for improvement and propose refinements to
    enhance the quality of EOP predictions.
  • Study the impact of geophysical and meteorological parameters on EOP prediction: The group seeks to investigate the influence of various geophysical and meteorological parameters on the prediction of Earth orientation parameters. By analyzing their correlations and dependencies, the study group aims to enhance the understanding of how these parameters affect EOP predictions, enabling more accurate
    and robust forecasting models.
  • Explore effective angular momentum prediction: The study group will focus on studying and improving the prediction of effective angular momentum, which plays a critical role in Earth’s orientation dynamics. By developing advanced prediction models and leveraging machine learning techniques, the group aims to enhance the accuracy and reliability of effective angular momentum predictions, contributing to
    improved EOP forecasts.
  • Study machine learning methods to derive the physical meaning and dependency structure between EOP and other parameters: The study group aims to investigate machine learning methods specifically designed to extract the physical meaning and underlying dependency structure between EOP and other geophysical and meteorological parameters. By applying these methods, the group seeks to uncover valuable insights and enhance the interpretability of AI-driven models, facilitating a deeper understanding of EOP dynamics.
  • Develop hybrid models to improve EOP prediction: The study group aims to
    develop hybrid prediction models that combine traditional methods with AI techniques.
    By integrating the strengths of different approaches, such as data-driven machine learn-
    ing algorithms and physical models, the group seeks to enhance the accuracy, robust-
    ness, and reliability of EOP predictions, providing more comprehensive forecasting
    capabilities.
  • Improve EOP theory using AI: By leveraging AI methodologies, the study group
    aims to advance the theoretical understanding of Earth orientation parameters. Through
    the analysis of complex datasets and the application of AI-driven techniques, the group
    seeks to refine and expand EOP theories, enabling a deeper comprehension of the un-
    derlying dynamics and factors influencing Earth’s orientation.
  • Conduct sensitivity analysis of EOP prediction models: The study group will
    perform sensitivity analyses to evaluate the stability and robustness of EOP prediction
    models. By systematically assessing the impact of various input parameters and as-
    sumptions, the group aims to identify critical factors that influence the accuracy and
    reliability of predictions, providing insights for model refinement and validation.
  • Explore commercial applications, such as satellite orbit determination and real-time applications: The study group aims to investigate the commercial applications of improved EOP predictions, particularly in areas such as satellite orbit determination and real-time applications. By understanding the impact of accurate EOP predictions on these domains, the group seeks to contribute to the development of innovative solutions and technologies that can enhance satellite operations and realtime positioning systems.
  • Identify the best prediction model for each Earth Orientation Parameter: The study group aims to evaluate and compare different prediction models for each Earth Orientation Parameter. By conducting comprehensive assessments and benchmarking exercises, the group seeks to identify the most accurate and reliable prediction
    methods for specific parameters, allowing for tailored approaches and optimized EOP predictions.
  • Foster knowledge sharing and collaboration: The study group will actively promote knowledge sharing, collaboration, and exchange of ideas among researchers, industry professionals, and policymakers. Through workshops, conferences, and publications, the group aims to facilitate discussions and interactions that  accelerate advancements in EOP prediction, promoting the adoption of AI techniques and fostering innovation in the field.

Members

  1. Dimitrios Ampatzidis (International Hellenic University, Thessaloniki, Greece, )
  2. Alireza A. Ardalan (University of Tehran, Tehran, Iran, )
  3. Esther Azcue (National Geographic Institute (IGN), Spain, )
  4. Jan Becker (Federal Agency for Cartography and Geodesy (BKG), Frankfurt, Germany, Jan.Becker@bkg.bund.de)
  5. Santiago Belda (University of Alicante, Applied Mathematics / Space Geodesy Group, Alicante, Spain, )
  6. Liliane Biskupek (Leibniz University Hannover, Hanover, Germany, )
  7. Sharyl Byram (United States Naval Observatory, USA, )
  8. Maria Davis (Lead Scientist at the IERS Rapid Service / Prediction Center at the U.S. Naval Observatory, )
  9. Robert Dill (GFZ, Potsdam, Germany, )
  10. Sujata Dhar (GFZ, Potsdam, Germany, )
  11. Sonia Guessoum (University of Alicante, Applied Mathematics / Space Geodesy Group, Alicante, Spain, )
  12. Junyang Gou (ETH Zürich, Zürich, Switzerland, )
  13. Dzana Halilovic (Federal Agency for Cartography and Geodesy (BKG), Frankfurt, Germany, Dzana.Halilovic@bkg.bund.de)
  14. Ibnu Nurul Huda (Badan Riset dan Inovasi Nasional (BRIN), Jakarta Pusat, Indonesia, )
  15. Mostafa Kiani Shahvandi (ETH Zürich, Zürich, Switzerland, )
  16. Qiaoli Kong (Shandong University of Science and Technology, Shandong, China, )
  17. Tomasz Kur (Institute of Geodesy and Geoinformatics, Wroclaw University of Environmental and Life Sciences, Wroclaw, Poland, )
  18. Arnab Laha (Indian Institute of Technology Kanpur, Kanpur, India, )
  19. Marcin Ligas (AGH University of Science and Technology, Krakow, Poland, )
  20. Zhao Li (GNSS center, Wuhan University, Wuhan, )
  21. Maciej Michalczak (AGH University of Science and Technology, Krakow, Poland, )
  22. Wei Miao (Shanghai Astronomical Observatory(SHAO), Chinese Academy of Sciences(CAS), Shanghai, China, )
  23. Sadegh Modiri (Federal Agency for Cartography and Geodesy (BKG), Frankfurt, Germany, Sadegh.Modiri@bkg.bund.de)
  24. Mariana Santos Moreira (Esta¸cao RAEGE de Santa Maria, Associa¸cao RAEGE A¸cores Santa Maria Azores, Portugal Atlantic International Research Centre, Terceira Azores, Portugal, )
  25. Johnson Oguntuase (University of Southern Mississippi, USA, )
  26. Jessica Page (United States Naval Observatory, USA, )
  27. Jihye Park (Oregon State University, Corvallis, USA, )
  28. Xanthos Papanikolaou (National Technical University of Athens, Athens, Greece, )
  29. Aleksander Partyka (Centrum Badań Kosmicznych Polskiej Akademii Nauk (CBK PAN), Warsaw, Poland, )
  30. Victor Puente (National Geographic Institute (IGN), Spain, )
  31. Haibo Que (China, )
  32. Shrishail Raut (GFZ, Potsdam, Germany, )
  33. Ole Roggenbuck (Federal Agency for Cartography and Geodesy (BKG), Frankfurt, Germany, Ole.Roggenbuck@bkg.bund.de)
  34. Harald Schuh (GFZ, Potsdam, Germany, )
  35. David Schunck (University of Tasmania, Hobart, Australia, )
  36. Yi Shen (Xinyang Normal University, China, )
  37. Sahayan Shirafkan (University of Tehran, Tehran, Iran, )
  38. Justyna Sliwinska (Centrum Badań Kosmicznych Polskiej Akademii Nauk (CBK PAN), Warsaw, Poland, )
  39. Robert Galatiya Suya (University of Nottingham, Nottingham, UK, )
  40. Lin Wang (Federal Agency for Cartography and Geodesy (BKG), Frankfurt, Germany, Lin.Wang@bkg.bund.de)
  41. Qianxin Wang (China University of Mining and Technology, China, )
  42. Na Wei (GNSS center, Wuhan University, Wuhan, )
  43. Danso Sampson Williams (University of Mines and Technology, Ghana, , Ghana)
  44. Yuanwei Wu (National Time Service Center (NTSC) of Chinese Academy of Sciences (CAS), Xi’an, China, )
  45. Yan Kailing (China, )
  46. Dongshan Yin (National Time Service Center (NTSC) of Chinese Academy of Sciences (CAS), Xian, China, )
  47. Kehao Yu (GNSS Research Center, Wuhan University, China, )
  48. Xueqing Xu (Shanghai Astronomical Observatory(SHAO), Chinese Academy of Sciences(CAS), Shanghai, China, )
  49. Zhongkai Zhang (Henan University, China, )