CNN-LSTM-based compound control for micro-vibration in inter-satellite optical communication systems

Area of Science:

• Space engineering
• Optical communications
• Control systems

Background:

• Micro-vibrations on satellite platforms cause pointing errors, degrading inter-satellite wireless optical communication (IsOWC) system stability and reliability.
• External perturbations and structural resonances are primary sources of these detrimental micro-vibrations.

Purpose of the Study:

• To develop and evaluate a novel compound control scheme for mitigating micro-vibrations in IsOWC systems.
• To enhance the pointing accuracy and overall performance of IsOWC by improving vibration suppression.

Main Methods:

• Implemented a compound control scheme integrating a convolutional long short-term memory (CNN-LSTM) network for predictive feedforward compensation.
• Combined CNN-LSTM prediction with a proportional-integral-derivative (PID) controller for enhanced micro-vibration suppression.
• Utilized the CNN-LSTM network to predict micro-vibration signals for proactive compensation.

Main Results:

• The proposed compound control scheme significantly reduced vibration peak amplitude by up to 83% compared to traditional PID control.
• Demonstrated substantial improvement in the stability and reliability of IsOWC systems.
• Validated the effectiveness of CNN-LSTM predictive feedforward compensation in micro-vibration suppression.

Conclusions:

• The proposed CNN-LSTM and PID compound control scheme offers a superior solution for micro-vibration suppression in IsOWC systems.
• This approach markedly enhances the stability and communication performance, crucial for reliable inter-satellite optical links.
• The predictive capabilities of CNN-LSTM networks are vital for advanced control strategies in space-based communication systems.