Research on satellite orbit prediction technology based on LSTM
Received date: 2023-12-09
Online published: 2024-11-16
Copyright
orbit prediction is crucial for assessing the likelihood of collisions between space objects and for better managing the near-earth space environment. Traditional orbit prediction methods rely on physical dynamic models, which necessitate the modeling of complex space environments and space objects. In reality, the limited understanding of many non-gravitational perturbations restricts the accuracy of orbit predictions. Considering the limitations of traditional orbit prediction methods, we proposed a technique to predict orbits based on the long short-term memory (LSTM) network. This approach leveraged a series of convolutions to extract features from the satellite orbit data over time, uncovering the underlying operational patterns. Experimental results indicated that this method improved the accuracy of satellite orbit predictions and provided a theoretical foundation for improving space situational awareness capabilities.
WU Xiaohe , ZHANG Kun , YUAN Shuai . Research on satellite orbit prediction technology based on LSTM[J]. Journal of Cybersecurity, 2024 , 2(4) : 18 -28 . DOI: 10.20172/j.issn.2097-3136.240402
表 1 Norad ID 44715卫星TLE第一行各字段含义Table 1 Norad ID 44715 Satellite TLE first line field meanings |
| 列号 | 描述 | 补充说明 |
| 01 | 行号 | 表示这是第一行 |
| 03-07 | 卫星的国际编号 | — |
| 08 | 卫星类别 | U代表该卫星非密,C 和S表示保密,仅限NORAD使用 |
| 10-11 | 发射时间 | 卫星发射年份的最后两位 |
| 12-14 | 当年发射数据 | 示例表示:2019年第74次发射 |
| 15-17 | 发射卫星个数 | A表示是第一个,如果一次发射多颗卫星,使用26个英文字母排序;若超过26,则使用两位字母,如AA、AB、AC编号 |
| 19-20 | TLE的历元年 | 历元年份的最后两位 |
| 21-32 | TLE的历元时刻 | 用十进制小数表示一年中的第几日和日中的小数部分 |
| 34-43 | 平均运动的一阶导数 | 用来计算每一天平均运动的变化带来的轨道漂移,提供给轨道计算软件预测卫星的位置,可以用来校准卫星的位置 |
| 45-52 | 平均运动的二阶导数 | 采用十进制小数,用来计算每一天平均运动的变化带来的轨道漂移,提供给轨道计算软件预测卫星的位置 |
| 54-61 | B* Drag Term(B*阻力项) | 影响低轨卫星运行的关键因素 |
| 63 | 星历类型 | 美国空军空间指挥中心内部使用标识为1,美国空军空间指挥中心以外公开使用标识为0 |
| 65-68 | 星历编号 | TLE数据按新发现卫星的先后顺序的编号 |
| 69 | 校验码 | 用于精确纠正误差 |
表 2 Norad ID 44715卫星TLE第二行各字段含义Table 2 Norad ID 44715 Satellite TLE second line field meanings |
| 列号 | 描述 | 补充说明 |
| 01 | 行号 | 表示这是第二行 |
| 03-07 | 卫星国际编号 | — |
| 09-16 | 卫星轨道倾角 | 范围在0到180°之间 |
| 18-25 | 升交点赤经 | 春分点和卫星升交点之间的角度,范围在0到360°之间 |
| 27-33 | 偏心率 | 表示轨道椭圆的形状,对于低轨卫星,其偏心率范围在0到0.25之间 |
| 35-42 | 近地点角 | 单位为度 |
| 44-51 | 平近点角 | 单位为度 |
| 53-63 | 平均运动 | 每天绕地球的圈数 |
| 64-68 | 历元轨道数 | 发射以来飞行的圈数 |
| 69 | 校验码 | 用于精确纠正误差 |
表 3 所选卫星参数Table 3 Parameters of selected satellites |
| Norad ID | 发射时间 | 数据范围 |
| 44715 | 2019.11.11 | 2020.02.01-2024.06.30 |
| 44298 | 2019.05.24 | 2020.05.01-2024.07.31 |
| 44942 | 2020.01.07 | 2020.01.14-2024.04.02 |
| 47368 | 2021.01.20 | 2021.05.01-2024.07.31 |
| 46798 | 2020.10.24 | 2020.11.04-2024.07.31 |
表 4 模型详细参数Table 4 Detailed parameters of the model |
| 参数名称 | 参数设置 |
| 128 | |
| 优化函数 | Adam |
| 学习率 | |
| 损失函数 | MSE |
| 激活函数 | Relu |
| Epoch | 50 |
| 0.5 | |
| 64 | |
| 4 | |
| 20 |
表 5 不同模型的实验结果Table 5 Experimental results of different models |
| 模型 | MAE | RMSE | |||||
| X(km) | Y(km) | Z(km) | X(km) | Y(km) | Z(km) | ||
| RNN | |||||||
| Seq2Seq | |||||||
| LSTM | | ||||||
表 6 不同编号卫星在LSTM上的性能Table 6 Performance of different numbered satellites on LSTM |
| Norad ID | MAE | RMSE | |||||
| X(km) | Y (km) | Z(km) | X(km) | Y(km) | Z(km) | ||
| 44942 | |||||||
| 44298 | |||||||
| 47368 | |||||||
| 46798 | |||||||
| 1 |
薛文, 胡敏, 阮永井, 等. 基于TLE的Starlink星座第一阶段部署情况分析[J]. 中国空间科学技术, 2022, 42 (5): 24- 33.
XUE W, HU M, RUAN Y J, et al. Analysis of the first stage deployment of Starlink constellation based on TLE[J]. Chinese Space Science and Technology., 2022, 42 (5): 24- 33.
|
| 2 |
LIN C S,CHEN J Y. Using bidirectional long short-term memory to improve orbit prediction accuracy of starlink[J]. Available at SSRN:https://ssrn.com/abstract=4707240.
|
| 3 |
JADALA G,MEEDINTI G N,DELHIBABU R. Satellite Orbit prediction using a machine learning approach[C]//Workshops at the 5th International Conference on Applied Informatics (ICAI). 2022:28-46.
|
| 4 |
PUENTE C, SÁENZ-NUÑO M A, VILLA-MONTE A, et al. Satellite orbit prediction using big data and soft computing techniques to avoid space collisions[J]. Mathematics, 2021, 9 (17): 2040.
|
| 5 |
HAMIDIAN S, KOSARI A R. A PCA-SVM regression model for LEO space debris orbit prediction in an optical space-based space surveillance network[J]. IEEE Access, 2024, 12, 60447- 60466.
|
| 6 |
CHEN Y W,WANG K Z. Prediction of satellite time series data based on long short term memory-autoregressive integrated moving average model (LSTM-ARIMA) [C]//2019 IEEE 4th International Conference on Signal and Image Processing (ICSIP). IEEE,2019:308-312.
|
| 7 |
OSAMA A,RAAFAT M,ABDELGHAFAR S,et al. Deep learning approach to satellite collision avoidance using long short-term memory[C]//In Doctoral Symposium on Intelligence Enabled Research. Singapore:Springer Nature Singapore,2023:101-111.
|
| 8 |
CHEN S Y, JIN G, MA X Y. Satellite on-orbit anomaly detection method based on a dynamic threshold and causality pruning[J]. IEEE ACCESS, 2021, 9, 986751- 86758.
|
| 9 |
LI B, LIU L, SANG J Z. Tracklet-to-object Matching for climbing starlink satellites through recursive orbit determination and prediction[J]. Research in Astronomy and Astrophysics, 2022, 22 (11): 115010.
|
| 10 |
JUNG O, SEONG J, JUNG Y, et al. Recurrent neural network model to predict re-entry trajectories of uncontrolled space objects[J]. Advances in Space Research, 2021, 68 (6): 2515- 2529.
|
| 11 |
LEE B S,HWANG Y,KIM D W. Machine learning approach to initial orbit determination of unknown LEO satellites[C]//In 2018 SpaceOps Conference. 2018:2566.
|
| 12 |
RANAWEERA R N,FERNANDO T. Prediction of potentially hazardous asteroids using deep learning[C] //2022 2nd International Conference on Advanced Research in Computing (ICARC). IEEE,2022:31-36.
|
| 13 |
SALLEH N,AZMI N F M,YUHANIZ S S. An adaptation of deep learning technique in orbit propagation model using long short-term memory[C]//2021 International Conference on Electrical,Communication,and Computer Engineering (ICECCE). IEEE,2021:1-6.
|
| 14 |
LI B, HUANG J, FENG Y, et al. A machine learning-based approach for improved orbit predictions of LEO space debris with sparse tracking data from a single station[J]. IEEE Transactions on Aerospace and Electronic Systems, 2020, 56 (6): 4253- 4268.
|
| 15 |
PÉREZ I ,SAN-JUAN F J ,SAN-MARTÍN M ,et al. Application of computational intelligence in order to develop hybrid orbit propagation methods[J]. Mathematical Problems in Engineering,2013,2013(1):631628.
|
| 16 |
SAN-JUAN F J, SAN-MARTÍN M, PÉREZ I, et al. Hybrid perturbation methods based on statistical time series models[J]. Advances in Space Research, 2016, 57 (8): 1641- 1651.
|
| 17 |
SAN-JUAN F J, PÉREZ I, SAN-MARTÍN M, et al. Hybrid SGP4 orbit propagator[J]. Acta Astronautica, 2017, 137, 254- 260.
|
| 18 |
HUANG W B, TANG R, QU G Z, et al. An XGBoost-based method for improved orbit prediction with an orbit-separate modeling strategy[J]. IEEE Transactions on Aerospace and Electronic Systems, 2024, 60 (4): 4887- 4895.
|
| 19 |
LI B, ZHANG Y, HUANG J, et al. Improved orbit predictions using two-line elements through error pattern mining and transferring[J]. Acta Astronautica, 2021, 188, 405- 415.
|
| 20 |
ABAY R, BALAGE S, BROWN M, et al. Two-line element estimation using machine learning[J]. The Journal of the Astronautical Sciences, 2021, 68 (1): 1- 27.
|
| 21 |
HAO P, BAI X L. Improving orbit prediction accuracy through supervised machine learning[J]. Advances in Space Research, 2018, 61 (10): 2628- 2646.
|
| 22 |
HAO P, BAI X L. Exploring capability of support vector machine for improving satellite orbit prediction accuracy[J]. Journal of Aerospace Information Systems, 2018, 15 (6): 366- 381.
|
| 23 |
HAO P, BAI X L. Machine learning approach to improve satellite orbit prediction accuracy using publicly available data[J]. The Journal of the Astronautical Sciences, 2019, 67 (2): 1- 32.
|
| 24 |
HAO P, BAI X L. A medium-scale study of using machine learning fusion to improve TLE prediction precision without external information[J]. Acta Astronautica, 2023, 204, 477- 491.
|
| 25 |
HAO P,BAI X L. Limits of machine learning approach on improving orbit prediction accuracy using support vector machine[C]//Advanced Maui Optical and Space Surveillance (AMOS) Technologies Conference. 2017:5.
|
| 26 |
HAO P, BAI X L. Gaussian Processes for improving orbit prediction accuracy[J]. Acta Astronautica, 2019, 161, 44- 56.
|
| 27 |
HAO P, BAI X L. Fusion of a machine learning approach and classical orbit predictions[J]. Acta Astronautica, 2021, 184, 222- 240.
|
| 28 |
HAO P, BAI X L. Artificial neural network–based machine learning approach to improve orbit prediction accuracy[J]. Journal of Spacecraft and Rockets, 2018, 55 (5): 1248- 1260.
|
| 29 |
HAO P, BAI X L. Comparative evaluation of three machine learning algorithms on improving orbit prediction accuracy[J]. Astrodynamics, 2019, 3 (4): 325- 343.
|
| 30 |
MIN Z, HU Z B, HU Y Y, et al. Improvement of orbit prediction accuracy using extreme gradient boosting and principal component analysis[J]. Open Astronomy, 2022, 31 (1): 229- 243.
|
| 31 |
GRAVES A. Long short-term memory[J]. Supervised sequence labelling with recurrent neural networks,2012:37-45.
|
| 32 |
SHIN Y J,PARK E J,WOO S S,et al. Selective tensorized multi-layer LSTM for orbit prediction[C]//In Proceedings of the 31st ACM International Conference on Information & Knowledge Management. 2022:3495-3504.
|
| 33 |
HONG W X,CHEN M T,GAO D,et al. Medium-and long-term orbit prediction of satellite based on LSTNet[C]//2023 15th International Conference on Computer and Automation Engineering (ICCAE). IEEE,2023:534-540.
|
| 34 |
OSAMA A,RAAFAT M,DARWISH A,et al. Satellite orbit prediction based on recurrent neural network using two line elements[C]//2022 5th International Conference on Computing and Informatics (ICCI). IEEE,2022:298-302.
|
| 35 |
ZHOU X Y, QIAO D, LI X Y. Neural network-based method for orbit uncertainty propagation and estimation[J]. IEEE Transactions on Aerospace and Electronic Systems, 2023, 60 (1): 1176- 1193.
|
| 36 |
THAMMAWICHAI M, LUANGWILAI T. Data-driven satellite orbit prediction using two-line elements[J]. Astronomy and Computing, 2024, 46, 100782.
|
| 37 |
REN H L,CHEN X L,GUAN B,et al. Research on satellite orbit prediction based on neural network algorithm[C]//Proceedings of the 2019 3rd High Performance Computing and Cluster Technologies Conference. 2019:267-273.
|
| 38 |
ChANG S Y,WU H C,SOTIROPOULOS F,et al. Robust satellite-orbit prediction using artificial neural network based on levenberg-marquardt algorithm[C]//In 2022 International Wireless Communications and Mobile Computing (IWCMC). IEEE,2022:1329-1334.
|
| 39 |
VASWANI A,SHAZEER N,PARMAR N,et al. Attention is all you need[C]//31st Conference on Neural Information Processing Systems (NIPS 2017),2017.
|
| 40 |
LAI G K,CHANG W C,YANG Y M,et al. Modeling long-and short-term temporal patterns with deep neural networks[C]//The 41st international ACM SIGIR conference on research & development in information retrieval. 2018:95-104.
|
| 41 |
SPACE-TRACK[EB/OL]. (2005-03-31)[2024-10-15]https://www.spacetrack.org.
|
/
| 〈 |
|
〉 |