پیش‌بینی بردار حالت مداری ماهواره با استفاده از سری‌های زمانی و شبکه‌های عصبی

نوع مقاله: مقالة‌ تحقیقی‌ (پژوهشی‌)

نویسنده

گروه مهندسی برق، دانشکدة فنی و مهندسی، دانشگاه صنعتی قوچان

چکیده

در این مقاله، روشی نوین برای پیش ­بینی موقعیت مداری ماهواره با استفاده از سری­ های زمانی و شبکه‌های عصبی معرفی شده است. در این روش، بر خلاف روش­های معمول پیش ­بینی مدار، از قوانین کپلر استفاده نشده و از قدرت پیش­ بینی سری ­های زمانی در شبکه ­های عصبی برای پیش­ بینی موقعیت مداری استفاده شده است. مهمترین مزیت روش پیشنهادی نسبت به روش­ های موجود، در استفاده از داده­های واقعی است. چرا که روش ­های موجود عموما با ساده سازی روابط و نیز حذف برخی از اغتشاشات معمولاً دارای خطا بوده و استفاده از معادلات بازگشتی نیز به‌طور افزاینده­ای این خطا را افزایش می­ دهد. در دسترس ترین داده واقعی، TLE بوده و دقت آنها نیز در پژوهش­ های مختلف به اثبات رسیده است. لذا در روش پیشنهادی استفاده از این داده­ ها در دستور کار قرار گرفته است. نتایج شبیه ­سازی و مقایسه این روش با الگوریتم SGP4 و داده­ های واقعی، نشان از کارآمدی روش پیشنهادی دارد.

کلیدواژه‌ها


عنوان مقاله [English]

Predicting Orbital State Vector of Satellites Using Time-Series Neural Networks

نویسنده [English]

  • Danial Boostan
The Quchan University of Technology
چکیده [English]

Prediction of satellite orbital position is a critical requirement for all satellite ground stations. In this paper, a new viewpoint for predicting orbital position of satellites is presented. In contrast to traditional methods which are based on Kepler's law, the proposed method, is solely based on past observation of a given satellite. In contrast to traditional algorithms which have neglected some of the orbital perturbations, the most important feature of this method is considering all orbital perturbations by using real data. TLEs (Two Line Element sets) are the most available real data and are used in this research as the main data source. Using the capability of neural networks for time series prediction over available data, results in a fast and accurate orbital position predictor. The comparison between the output of our proposed method, SPG4 (Simplified General Perturbation version 4) propagator and real orbital position of a given satellite, shows the effectiveness of this algorithm.

کلیدواژه‌ها [English]

  • time series prediction
  • Artificial Neural Networks
  • TLE (Two Line Elements)
  • satellite orbital position

[1]   Navabi, M. and Hamrah, R., Modeling of Space Objects Propagation, Prediction of Closest Approaches among Satellites, and Assessment of Maximum Collision Probability, Journal of Space Science and Technology (JSST), Vol. 6, No. 1, 2013, 57-67 (in Persian).

[2]   Neta, B., Partial List of Orbit Propagators, Naval Postgraduate School.

[3]   Miura, N.Z., Comparison and Design of Simplified General Perturbation Models (SGP4) and Code for NASA Johnson Space Center, (Thesis M. Sc.) 2009.

[4]   Chao, C.C., Warner, L.F.,  Cox, J., Thompson, R.C., Starchville, T.F., Cook, J.W. and Woodburn, J., “IV&V of Three Astrodynamics Functions of the Satellite Tool Kit, AAS/AIAA Astrodyn,” Spec. Conf., 2000, pp. 70–81.

[5]   Aorpimai, M., Malayavej, V. and Navakitkanok, P., High-fidelity Orbit Propagator for Precise Antenna Pointing in LEO Satellite Operation, 20th Asia-Pacific Conf. Commun., 2014, pp. 223–226..

[6]   Fang, L. and Nagarajan, N., Neural Network Based Orbit Propagation for Small Satellite Missions,  Small Satell. Conf., 2004, pp. 1–9. http://digitalcommons.usu.edu/smallsat/2004/All2004/47/.

[7]   Greene, M.R. and Zee, R.E., “Increasing the Accuracy of Orbital Position Information from NORAD SGP4 Using Intermittent GPS Readings,” Small Satell. Conf., 2009.

[8]   Seppänen, M., Ala-Luhtala, J., Piché, R., Martikainen, S. and Ali-Löytty, S., Autonomous Prediction of GPS and GLONASS Satellite Orbits, Navig. J. Inst. Navig, Vol. 59, 2012, pp.119–134.

[9]   Amiri, M.A., Gazerpour, A, H., Roghangir, S. A. R., Orbit Determination Via a Deeply- Coupled UKF/GNSS Filter, Journal of Space Science and Technology (JSST), Vol. 9, No. 3, 2016, pp. 13-26, (In Persian).

[10] Torabi, P. and Naghash, A., Determining orbital element on Earth Observation Repeat-Ground-Track orbit, Journal of Space Science and Technology (JSST), Vol. 9, No. 2, 2016, pp. 77-83, (In Persian).

[11] Bennett, J., Sang, J.,  Smith, C.,  Zhang, K., Improving Low-Earth Orbit Predictions Using Two-line Element Data with Bias Correction, Advanced Maui Optical and Space Surveillance Technologies Conference, 2012.

[12] Peng, H. and Bai, X., Improving Orbit Prediction Accuracy Through Supervised Machine Learning, Advances in Space Research, Vol 10, No. 9, 2018, pp. 2628-2646

[13] D. Bustan, N. Pariz, S.K. Hosseini Sani, Intelligent Satellite Orbit Prediction Based on Time Series Analysis, Journal of Space Science and Technology (JSST), Vol 7, No 3, 2014, 43–49 (In Persian)

[14] Shamlu, F. and Naghash, A. “Satellite Orbit Prediction Through Observation Data and the Artificial Neural Networks,” Journal of Space Science and Technology (JSST), Vol 10, No 2, 2017, pp. 1-8 (In Persian)

[15]https://spaceflight.nasa.gov/realdata/sightings/SSapplications/Post/JavaSSOP/SSOP_Help/tle_def.html

[16] H.D. Curtis, Orbital Mechanics for Engineering Students, 2014.

[17] Vallado, D.A., Crawford, P., Hujsak, R., Kelso, T.S., Revisiting Spacetrack Report #3, AIAA, 2006.

[18] Samarasinghe, S., Neural Networks for Applied Sciences and Engineering: From Fundamentals to Complex Pattern Recognition, Auerbach Publications, Boston, 2006.

[19] Beale, M.H., Demuth, H.B., Hagan, M.T. and Chen, Q., Neural network toolboxTM 8, Natick, Massachusetts, 2014.

[20] Kelso, T.S., Validation of SGP4 and IS-GPS-200D Against GPS Precision Ephemerides, Adv. Astronaut. Sci., 2007, pp. 427–440.

[21] Wikipedia Contributors, “Two-line element set,” Wikipedia, The Free Encyclopedia, https:// en.wikipedia.org/w/index.php?title=Two-line_ element_set&oldid=790226837

[22] Kelso, T.S., No Title, (n.d.). Celestrak.com.

[23] Han, J. and Kamber, M., Data Mining: Concepts and Techniques, 2006,

[24] Lourakis, M.I.A., A Brief Description of the Levenberg-Marquardt Algorithm Implemented By Levmar, Foundation of Research and Technology, 2005, pp. 1-6.