An in-orbit correction method based on CNN for the figure errors and component misalignments of TMA telescope

Bingdao Li, Xiaofang Zhang*, Yun Gu, Xinqi Hu

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

To realize the fast and simple in-orbit aberration correction of TMA telescope, an aberration correction method based on Convolutional Neural Network (CNN) is proposed. CNN is trained to establish the relationship between the defocus point spread function and the misalignments of the secondary mirror. The wavefront aberration caused by the figure errors of the primary mirror and the misalignments of the secondary mirror and the tertiary mirror can be compensated by adjusting the secondary mirror according to the outputs of the well-trained CNN (named as Cor-Net). This method can correct the system aberration quickly and the RMS of the system wavefront aberration is reduced from about 1.5 λ to 0.1 λ by only three correction cycles.

Original languageEnglish
Title of host publicationOptical Design and Testing XII
EditorsYongtian Wang, Tina E. Kidger, Rengmao Wu
PublisherSPIE
ISBN (Electronic)9781510656963
DOIs
Publication statusPublished - 2023
EventOptical Design and Testing XII 2022 - Virtual, Online, China
Duration: 5 Dec 202211 Dec 2022

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume12315
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceOptical Design and Testing XII 2022
Country/TerritoryChina
CityVirtual, Online
Period5/12/2211/12/22

Keywords

  • CNN
  • In-orbit correction
  • defocus point spread function
  • figure errors
  • misalignments

Fingerprint

Dive into the research topics of 'An in-orbit correction method based on CNN for the figure errors and component misalignments of TMA telescope'. Together they form a unique fingerprint.

Cite this