Satellite attitude tracking control of moving targets combining deep reinforcement learning and predefined-time stability considering energy optimization
•An attitude tracking control method combining deep reinforcement learning and traditional satellite attitude controller for moving targets tracking and observation is proposed.•Integrate LSTM into TD3 to learn the moving state of the target from its image positions and obtain desired attitude in re...
Uložené v:
| Vydané v: | Advances in space research Ročník 69; číslo 5; s. 2182 - 2196 |
|---|---|
| Hlavní autori: | , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Elsevier B.V
01.03.2022
|
| Predmet: | |
| ISSN: | 0273-1177, 1879-1948 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Shrnutí: | •An attitude tracking control method combining deep reinforcement learning and traditional satellite attitude controller for moving targets tracking and observation is proposed.•Integrate LSTM into TD3 to learn the moving state of the target from its image positions and obtain desired attitude in real time.•Energy optimization is considered in the design of the reward function of the TD3, which promotes the TD3 to learn a strategy of generating the desired attitude that satellite maneuvers to with less energy.•An adaptive backstepping controller with predefined-time stability is designed to ensure that the satellite attitude reaches the desired attitude within a predefined decision period in the presence of the external disturbance and uncertain inertia properties.•A simulation system of moving target tracking and observation is established, and the results indicate that our approach is superior in terms of tracking ability and energy consumption.
Space-based moving targets tracking and observation facilitates target recognition and analysis of target characteristics, but the ability of satellite attitude tracking control needs to be improved, especially considering the energy optimization for long-time tracking. An attitude tracking control method combining deep reinforcement learning and predefined-time stability is proposed, which not only improves the autonomous decision-making ability but also ensures the reliability of the satellite attitude control. The long short-term memory network is integrated into the twin delayed deep deterministic policy gradient algorithm to learn the moving state of the target from its image positions as the input to generate the desired attitude in real time, and energy optimization is considered in the design of the reward function. Then an adaptive backstepping controller is designed to achieve predefined-time stability in the presence of the external disturbance and uncertain inertia properties, which ensures that the satellite attitude is controlled to the desired value within a predefined decision period. Finally, a simulation system of moving target tracking is established, and the results indicate that our approach is superior in terms of tracking ability and energy consumption. |
|---|---|
| ISSN: | 0273-1177 1879-1948 |
| DOI: | 10.1016/j.asr.2021.12.014 |