# Generating Low-Earth Orbit Satellite Attitude Maneuver Profiles Using Deep Neural Networks

## Abstract

**:**

## 1. Introduction

- This paper proposes a ground-based technique for generating optimal reference profiles for maneuvers between target pointings of LEO observation satellites.
- The proposed technique predicts the quaternions of the maneuver through data-based learning and uses the predicted value as an initial guess for optimization to generate the maneuver profile.
- To demonstrate the performance of the proposed technique, the error of the starting point of the target pointing according to various maneuver angles, start and end angular rates are analyzed, and its performance is compared with the existing technique.

## 2. Background

#### 2.1. LEO Satellite Attitude Guidance Profile

#### 2.2. LEO Satellite Attitude Guidance Profile Constraints

## 3. Proposed Technique for Generating Attitude Reference Profiles

_{1}to t

_{7}are given as

_{7}were generated onboard the satellite using the uploaded acceleration and time data. The commands for satellite attitude maneuvering were provided in quaternion form instead of Euler angles to avoid gimbal locks and other issues. The attitude profile for the maneuvering segment was converted into a quaternion form for training using a direction cosine matrix (DCM).

_{8}was determined at the start of the mission according to the mission design requirements for the satellite, and it was defined as the difference between the user-controlled maneuver time and the time interval from ${t}_{1}$ to ${t}_{7}$. To account for the size of the uploaded data, the same time (${t}_{1}$ to ${t}_{8}$) values were used for the roll, pitch, and yaw axes. In other words, all three axes had the same timeline, and only the acceleration values were generated differently.

_{8}as follows:

^{−4}and a rate error of 10

^{−3}°/s were used for optimization. The process for generating the maneuver reference profile for LEO satellites is shown in Figure 5.

## 4. DNN Training Results

## 5. Case Studies

## 6. Conclusions

## Funding

## Data Availability Statement

## Conflicts of Interest

## References

- Geng, L.; Xiao, D.; Wang, Q.; Zhang, T.; Song, J. Two-Stage Identification Method for Attitude-Control Models of On-Orbit Satellites. Tsinghua Sci. Technol.
**2020**, 15, 319–324. [Google Scholar] [CrossRef] - Wahballah, W.A.; Eltohamy, F.; Bazan, T.M. Influence of Attitude Parameters on Image Quality of Very High-Resolution Satellite Telescopes. IEEE Trans. Aerosp. Electron. Syst.
**2021**, 57, 1177–1183. [Google Scholar] [CrossRef] - Xiao, B.; Yin, S.; Wu, L. A Structure Simple Controller for Satellite Attitude Tracking Maneuver. IEEE Trans. Ind. Electron.
**2017**, 64, 1436–1446. [Google Scholar] [CrossRef] - Show, L.-L.; Juang, J.-C.; Jan, Y.-W. An LMI-Based Nonlinear Attitude Control Approach. IEEE Trans. Control Syst. Technol.
**2003**, 11, 73–83. [Google Scholar] [CrossRef] - Zhao, L.; Zhao, K.; Hao, Y.; Zhang, X.; He, Y. The Attitude Control Algorithm of Agile Optical Satellite Oriented to Nonparallel-Ground-Track-Imaging. IEEE Access
**2019**, 7, 164362–164373. [Google Scholar] [CrossRef] - Steyn, W.H. Near-Minimum-Time Eigenaxis Rotation Maneuvers Using Reaction Wheels. J. Guid. Control Dyn.
**1995**, 18, 1184–1189. [Google Scholar] [CrossRef] - Wie, B.; Bailey, D.; Heiberg, C. Rapid Multitarget Acquisition and Pointing Control of Agile Spacecraft. J. Guid. Control Dyn.
**2002**, 25, 96–104. [Google Scholar] [CrossRef] - Wie, B.; Liu, Q. Comparison between Robustified Feedforward and Feedback for Achieving Parameter Robustness. J. Guid. Control Dyn.
**1992**, 15, 935–943. [Google Scholar] [CrossRef] - Singer, N.C.; Seering, W.P. Preshaping Command Inputs to Reduce System Vibration. Trans. ASME
**1990**, 112, 76–82. [Google Scholar] [CrossRef] - Bang, H.; Junkins, J.L. Lyapunov Optimal Control Law for Flexible Space Structure Maneuver and Vibration Control. J. Astronaut. Sci.
**1993**, 41, 91–118. [Google Scholar] - Jardak, N.; Jault, Q. The Potential of LEO Satellite-Based Opportunistic Navigation for High Dynamic Applications. Sensors
**2022**, 22, 2541. [Google Scholar] [CrossRef] [PubMed] - Wang, S.; Zhao, L.; Cheng, J.; Zhou, J.; Wang, Y. Task Scheduling and Attitude Planning for Agile Earth Observation Satellite with Intensive Tasks. Aerosp. Sci. Technol.
**2019**, 90, 23–33. [Google Scholar] [CrossRef] - Cui, P.; Yan, N. Research on Modeling of the Agile Satellite Using a Single Gimbal Magnetically Suspended CMG and the Disturbance Feedforward Compensation for Rotors. Sensors
**2012**, 12, 16964–16987. [Google Scholar] [CrossRef] [PubMed] - Zhang, X.; Zhang, X.; Lu, Z.; Liao, W. Optimal Path Planning-Based Finite-Time Control for Agile CubeSat Attitude Maneuver. IEEE Access
**2019**, 7, 102186–102198. [Google Scholar] [CrossRef] - Lafontaine, J.; Peuvedic, C.L. Autonomous Generation of Guidance Profiles for Constrained, Minimum-Time, Large-Angle Attitude Manoeuvres. In Proceedings of the AAS/AIAA Space Flight Mechanics Meeting, AAS, Cooper Mountain, CO, USA, 23–27 January 2005. [Google Scholar]
- Wu, S.; Liu, Y.; Radice, G.; Tan, S. Autonomous Pointing Control of a Large Satellite Antenna Subject to Parametric Uncertainty. Sensors
**2017**, 17, 560. [Google Scholar] [CrossRef] [PubMed] - Wu, S.; Liu, L.; Wang, R.; Wu, Z. On-Orbit High-Accuracy Pointing Control for Large Mesh Deployable Antenna. J. Astronaut.
**2015**, 36, 1140–1147. [Google Scholar] - Ahn, H.S.; Jung, O.; Choi, S.; Son, J.H.; Chung, D.; Kim, G. An Optimal Satellite Antenna Profile Using Reinforcement Learning. IEEE Trans. Syst. Man Cybern. Part C
**2011**, 41, 393–406. [Google Scholar] [CrossRef] - Chávez-Moreno, R.; Tang, Y.; Hernández, J.C.; Ji, H. Contracting Angular Velocity Observer for Small Satellites. IEEE Trans. Aerosp. Electron. Syst.
**2018**, 54, 2762–2775. [Google Scholar] [CrossRef] - Stollenwerk, T.; Michaud, V.; Lobe, E.; Picard, M.; Basermann, A.; Botter, T. Agile Earth Observation Satellite Scheduling with a Quantum Annealer. IEEE Trans. Aerosp. Electron. Syst.
**2021**, 57, 3520–3528. [Google Scholar] [CrossRef] - Kristiansen, R.; Nicklasson, P.J.; Gravdahl, J.T. Satellite Attitude Control by Quaternion-Based Backstepping. IEEE Trans. Control Syst. Technol.
**2009**, 17, 227–232. [Google Scholar] [CrossRef] - Ali, A.J.; Khalily, M.; Sattarzadeh, A.; Massoud, A.; Hasna, M.O.; Khattab, T.; Yurduseven, O.; Tafazolli, R. Power Budgeting of LEO Satellites: An Electrical Power System Design for 5G Missions. IEEE Access
**2021**, 9, 113258–113269. [Google Scholar] [CrossRef] - Li, K.-X.; You, L.; Wang, J.; Gao, X.; Tsinos, C.G.; Chatzinotas, S.; Ottersten, B. Downlink Transmit Design for Massive MIMO LEO Satellite Communications. IEEE Trans. Commun.
**2022**, 70, 1014–1028. [Google Scholar] [CrossRef] - Bahrami, S.; Namvar, M. Rigid Body Attitude Control with Delayed Attitude Measurement. IEEE Trans. Control Syst. Technol.
**2015**, 23, 1961–1969. [Google Scholar] [CrossRef] - Xiao, B.; Yin, S.; Kaynak, O. Attitude Stabilization Control of Flexible Satellites with High Accuracy: An Estimator-Based Approach. IEEE ASME Trans. Mechatron.
**2017**, 22, 349–358. [Google Scholar] [CrossRef] - Kumar, K.; Kumar, K.D. Line-of-sight pointing stability for “drifting” Satellites. IEEE Trans. Aerosp. Electron. Syst.
**1999**, 35, 504–510. [Google Scholar] [CrossRef] - Sun, C.; Jiang, H.; Zhang, J.; Tao, Y.; Li, B.; Zhao, C. Modeling and Calibrating the Ground-Surface Beam Pointing of GEO Satellite. IEEE Access
**2019**, 7, 121897–121906. [Google Scholar] [CrossRef] - Li, Y.; Wang, L. Human Activity Recognition Based on Residual Network and BiLSTM. Sensors
**2022**, 22, 635. [Google Scholar] [CrossRef] - Wang, Q.; Yu, Y.; Ahmed, H.O.A.; Darwish, M.; Nandi, A.K. Open-Circuit Fault Detection and Classification of Modular Multilevel Converters in High Voltage Direct Current Systems (MMC-HVDC) with Long Short-Term Memory (LSTM) Method. Sensors
**2021**, 21, 4159. [Google Scholar] [CrossRef] - Long, D.; Wen, X.; Zhang, W.; Wang, J. Recurrent Neural Network Based Robust Actuator and Sensor Fault Estimation for Satellite Attitude Control System. IEEE Access
**2020**, 8, 183165–183174. [Google Scholar] [CrossRef] - Peng, Y.; Song, D.; Qiu, L.; Wang, H.; He, X.; Liu, Q. Combined Prediction Model of Gas Concentration Based on Indicators Dynamic Optimization and Bi-LSTMs. Sensors
**2023**, 23, 2883. [Google Scholar] [CrossRef] - Greff, K.; Srivastava, R.K.; Koutník, J.; Steunebrink, B.R.; Schmidhuber, J. LSTM: A Search Space Odyssey. IEEE Trans. Neural Netw. Learn. Syst.
**2017**, 28, 2222–2232. [Google Scholar] [CrossRef] [PubMed]

**Figure 11.**Discontinuous quaternion errors at initial position of target pointing generated by proposed technique as a function of maneuver angle.

**Figure 12.**Discontinuous angular rate errors at initial position of target pointing generated by proposed technique as a function of maneuver angle.

**Figure 13.**Discontinuous quaternion errors at initial position of target pointing generated by proposed technique as a function of satellite rate difference between target points.

**Figure 14.**Discontinuous angular rate errors at initial position of target pointing generated by proposed technique as a function of satellite rate difference between target points.

**Figure 15.**Discontinuous quaternion error at initial position of target pointing as a function of tilt angle.

**Figure 16.**Discontinuous angle error at initial position of target pointing as a function of tilt angle.

**Table 1.**Parameters of proposed technique for generating the reference attitude between target pointings.

Parameter | |
---|---|

Roll | Maximum acceleration, Minimum acceleration for roll axis |

Pitch | Maximum acceleration, Minimum acceleration for pitch axis |

Yaw | Maximum acceleration, Minimum acceleration for yaw axis |

Time | Acceleration duration t _{1}, t_{2}, t_{3}, t_{4}, t_{5}, t_{6}, t_{7} |

Parameter | Number of Features | |
---|---|---|

Input features | Current point quaternion | 4 |

Satellite position vector | 3 | |

Satellite velocity vector | 3 | |

Relative angle difference to target point | 3 | |

Relative angular rate difference to target point | 3 | |

Output features | Next point quaternion | 4 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2023 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Yun, S.-T.
Generating Low-Earth Orbit Satellite Attitude Maneuver Profiles Using Deep Neural Networks. *Sensors* **2023**, *23*, 4650.
https://doi.org/10.3390/s23104650

**AMA Style**

Yun S-T.
Generating Low-Earth Orbit Satellite Attitude Maneuver Profiles Using Deep Neural Networks. *Sensors*. 2023; 23(10):4650.
https://doi.org/10.3390/s23104650

**Chicago/Turabian Style**

Yun, Seok-Teak.
2023. "Generating Low-Earth Orbit Satellite Attitude Maneuver Profiles Using Deep Neural Networks" *Sensors* 23, no. 10: 4650.
https://doi.org/10.3390/s23104650