مجله ی مهندسی مکانیک شریف

مجله ی مهندسی مکانیک شریف

تعیین وضعیت یک ماهواره ی زمین‌آهنگ مجهز به حسگرهای خورشید و زمین با استفاده از الگوریتم‌های هوش مصنوعی

نوع مقاله : مقاله پژوهشی

نویسندگان
فرید حاجی ویسه 1
حسین نجات پیشکناری 1
طالب عبداللهی 2
1 دانشکده‌ی مهندسی مکانیک، دانشگاه صنعتی شریف، تهران، ایران.
2 پژوهشگاه فضایی ایران، تهران، ایران
10.24200/j40.2026.68496.1761
چکیده
تعیین وضعیت، یکی از مؤلفه‌های حیاتی در طراحی و عملکرد سامانه‌های فضایی است. در پژوهش حاضر، چارچوبی نوین برای تخمین وضعیت یک ماهواره‌ی زمین‌آهنگ با ترکیب حسگرهای خورشید و زمین و استفاده از الگوریتم‌های هوش مصنوعی ارائه شده است. روش‌ جبریTRIAD برای پیش‌پردازش داده‌های خام حسگر استفاده شده است. در ادامه، مجموعه‌ای از الگوریتم‌های یادگیری ماشین و یادگیری عمیق شامل روش­های XGBoost، CatBoost، و شبکه‌ی عصبی چندلایه (MLP) برای تخمین کواترنیون‌های چرخش ارزیابی شده و نتایج نشان داده‌اند که استفاده‌ی هم‌زمان از چهار مدل MLP دو لایه‌ای با داده‌های پیش‌پردازش‌شده، همراه با نرمال‌سازی کواترنیون‌های خروجی، به میانگین خطای مطلق ۵۷/۰ درجه منجر می‌شود. رویکرد اخیر، تقریباً 10% برتری عملکردی را در مقایسه با الگوریتم TRIAD از خود نشان داده و دقت مناسبی برای کاربردهای عملی و اجرای بلادرنگ فراهم ساخته است.
کلیدواژه‌ها
تعیین وضعیت
ماهواره
هوش مصنوعی
حسگر خورشید
حسگر زمین.
موضوعات

عنوان مقاله English

Attitude Determination of Geosynchronous Satellites with Sun and Earth Sensors Using Artificial Intelligence Algorithms

نویسندگان English

Farid Hajiweisah 1
Hossein Nejat Pishkenari 1
Taleb Abdollahi 2
1 Faculty of Mechanical Engineering, Sharif University of Technology, Tehran, Iran
2 Iranian Space Research Center, Tehran, Iran
چکیده English

Attitude determination is a critical component in the design and operation of satellites. This study presents a novel framework for estimating the attitude of a geosynchronous satellite by combining Sun and Earth sensors with artificial intelligence algorithms. Using raw sensor data led to poor or unsatisfactory results, with the best prediction error exceeding 10 degrees; therefore, the raw data were augmented through a preprocessing step utilizing the TRIAD algebraic method. Several machine learning and deep learning algorithms, including XGBoost, CatBoost, and multilayer perceptron (MLP) networks, were then employed to estimate the satellite's rotation quaternions. The findings demonstrated that the simultaneous and parallel use of four identical two-layer MLP models, each predicting a single quaternion component from the preprocessed data, coupled with output normalization, yielded a mean absolute error of 0.57 degrees. This approach outperformed the TRIAD algorithm by approximately 10%, the best classical method evaluated, and its computation time was shown to be compatible with real-time onboard execution. This hybrid approach offers a cost-effective and reliable solution for implementing attitude determination systems in satellites.

کلیدواژه‌ها English

Attitude determination
Satellite
Artificial intelligence
Sun sensor
Earth sensor
1.  Orozco, M.L. and Giraldo, B.S., 2024. Attitude determination and control in small satellites: A review. IEEE Journal on Miniaturization for Air and Space Systems, 5(3), pp. 182-186. https://doi.org/10.1109/JMASS.2024.3402984.
2.  Fathi Jegarkandi, M. and Safaei, R., 2024. Fractional Order Sliding Mode Controller (FOSMC) Design for Attitude Control of a Satellite with Coupled Rigid-Flexible Structures Using Fractional Order Transfer Function. Sharif Journal of Mechanical Engineering (SJME), 40(2), pp. 3-14. https://doi.org/10.24200/j40.2023.62462.1684.
3.  Soken, H.E., Hajiyev, C. and Sakai, S.I., 2014. Robust Kalman filtering for small satellite attitude estimation in the presence of measurement faults. European Journal of Control, 20(2), pp. 64-72. https://doi.org/10.1016/j.ejcon.2013.12.002.
4.  Hajiyev, C. and Guler, D.C., 2017. Review on gyroless attitude determination methods for small satellites. Progress in Aerospace Sciences, 90, pp. 54-66. https://doi.org/10.1016/j.paerosci.2017.03.003.
5.  Dos Santos, G.H., Seman, L.O., Bezerra, E.A., Leithardt, V.R.Q., Mendes, A.S. and Stefenon, S.F., 2021. Static attitude determination using convolutional neural networks. Sensors, 21(19), p. 6419. https://doi.org/10.3390/S21196419.
6.  Ni, S. and Zhang, C., 2011. Attitude determination of nano satellite based on gyroscope, sun sensor and magnetometer. Procedia Engineering, 15, pp. 959-963. https://doi.org/10.1016/j.proeng.2011.08.177.
7.  Kikuya, Y., Iwasaki, Y., Yatsu, Y. and Matunaga, S., 2021. Attitude determination algorithm using Earth sensor images and image recognition. Transactions of the Japan Society for Aeronautical and Space Sciences, 64(2), pp. 82-90. https://doi.org/10.2322/TJSASS.64.82.
8.  Zhu, F., Qin, Y., Cai, Q., Chen, X. and Zhang, X., 2025. Combining star tracker and gyroscope for remote sensing satellite attitude determination: algorithm design and performance evaluation with real data. IEEE Transactions on Instrumentation and Measurement, 74, pp. 1-15. https://doi.org/10.1109/TIM.2025.3548235.
9.  Li, R., Feng, L., Wu, P., Deng, X. and Shi, H., 2025. A roll attitude determination method based on the jamming energy of GEO satellites and an LSTM neural network. Scientific Reports, 15(1), p. 17888. https://doi.org/10.1038/s41598-025-01174-0.
10. Nobahari, H., Ghanbarpour Asl, H. and farhad Abtahi, S.F., 2016. Investigation of Accuracy of Integrated Inertial-Radar Navigation Systems in Autolanding of Unmanned Air Vehicles. Sharif Journal of Mechanical Engineering (SJME), 32.3(2), pp. 53-62. https://dor.isc.ac/dor/20.1001.1.26764725.1395.323.2.7.8.
11. Sun, Z., Simo, J. and Gong, S., 2023. Satellite attitude identification and prediction based on neural network compensation, Space: Science & Technology, 3, p. 0009. https://doi.org/10.34133/SPACE.0009.
12. Quinlan, J.R., 1986. Induction of decision trees, Machine Learning, 1(1), pp. 81–106. https://doi.org/10.1007/BF00116251.
13. Breiman, L., 2001. Random forests, Machine Learning, 45(1), pp. 5–32. https://doi.org/10.1023/A:1010933404324.
14. Freund, Y. and Schapire, R.E., 1997. A decision-theoretic generalization of on-line learning and an application to boosting, Journal of Computer and System Sciences, 55(1), pp. 119–139. https://doi.org/10.1006/JCSS.1997.1504.
15. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A.V. and Gulin, A., 2018. CatBoost: unbiased boosting with categorical features. In Proceedings of the 32nd International Conference on Neural Information Processing Systems, pp. 6639-6649. https://dl.acm.org/doi/10.5555/3327757.3327770.
16. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q. and Liu, T.Y., 2017. Lightgbm: A highly efficient gradient boosting decision tree. In Proceedings of the 31st International Conference on Neural Information Processing Systems, pp. 3149-3157. https://dl.acm.org/doi/10.5555/3294996.3295074
17. Chen, T. and Guestrin, C., 2016. XGBoost: A Scalable Tree Boosting System. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pp. 785–794. https://doi.org/10.1145/2939672.2939785.
18. Rumelhart, D.E., Hinton, G.E. and     Williams, R.J., 1986. Learning representations by back-propagating errors, Nature, 323(6088), pp. 533–536. https://doi.org/10.1038/323533A0.
19. Berger, R.W., Bayles, D., Brown, R., Doyle, S., Kazemzadeh, A., Knowles, K., Moser, D., Rodgers, J., Saari, B., Stanley, D. and Grant, B., 2001. The RAD750TM - A Radiation Hardened PowerPCTM Processor for High Performance Spaceborne Applications. In IEEE Aerospace Conference Proceedings, Big Sky, MT, USA, 5, pp. 2263-2272. https://doi.org/10.1109/AERO.2001.931184.
20. Pasetti, A., 2002. Background on Attitude and Orbit Control Systems (AOCS). AOCS Framework Project, PnP Software. Available at: https://www.pnp-software.com/AocsFramework/AocsBackground.html [Accessed 9 Aug. 2026].