Co-phase errors sensing method for Golay3 telescope system via a transfer network.

Area of Science:

• Optical Engineering
• Astronomy
• Machine Learning

Background:

• Optical sparse-aperture systems require precise co-phase error detection and correction.
• Existing methods often necessitate additional optical components and can be time-consuming.

Purpose of the Study:

• To develop an efficient and accurate method for analyzing and compensating co-phase errors in Golay3 telescope systems.
• To establish an error compensation control system utilizing a deep learning approach.

Main Methods:

• A pre-trained neural network was fine-tuned to generate hash-like binary codes for image features.
• An index database was created linking image deep features to co-phase error values.
• A hierarchical deep search database was employed for rapid retrieval of error data.

Main Results:

• The developed system effectively detects and compensates for piston errors within [-5,5]λ and tilt errors within [-15,15]µrad.
• The method demonstrated high correction accuracy and a short training time.
• Simultaneous detection of both piston and tilt errors was achieved.

Conclusions:

• The proposed search framework offers a viable solution for co-phase error correction in optical sparse-aperture systems.
• This approach eliminates the need for extra optical components, offering a more streamlined and accurate correction system.