Build the Structure of WFSless AO System Through Deep Reinforcement Learning
Hu, K.1,2; Xu, Z.X.1; Yang, W.2; Xu, B.1
刊名IEEE PHOTONICS TECHNOLOGY LETTERS
2018-12-01
卷号30期号:23页码:2033-2036
关键词WFSless AO deep reinforcement learning DDPG CNN SPGD AOG
ISSN号1041-1135
DOI10.1109/LPT.2018.2874998
文献子类J
英文摘要We report on an aberration correction algorithm for a wavefront sensorless adaptive optics (WFSless AO) system based on deep reinforcement learning. First, it is verified that the reinforcement learning theory can be applied in our system. In addition, the deep deterministic policy gradient algorithm is introduced to build the control structure. After that, deep learning is used to deal with the messy raw images of far-field intensity distribution. We emphatically present how to design a feature extraction with the convolutional neural network in the control structure. To demonstrate the performance of this structure, some comparisons are made with the stochastic parallel gradient descent algorithm and the WFSless AO based on general modes algorithm. The results indicate that the correction speed of our method improves about 9 times and 2.5 times, respectively, for the similar correction effect.
WOS关键词ABERRATION CORRECTION ; ALGORITHM
语种英语
WOS记录号WOS:000451233300010
内容类型期刊论文
源URL[http://ir.ioe.ac.cn/handle/181551/9286]  
专题光电技术研究所_自适应光学技术研究室(八室)
作者单位1.Key Laboratory of Adaptive Optics, Chinese Academy of Sciences, Chengdu; 610209, China;
2.University of Chinese Academy of Sciences, Beijing; 100049, China
推荐引用方式
GB/T 7714
Hu, K.,Xu, Z.X.,Yang, W.,et al. Build the Structure of WFSless AO System Through Deep Reinforcement Learning[J]. IEEE PHOTONICS TECHNOLOGY LETTERS,2018,30(23):2033-2036.
APA Hu, K.,Xu, Z.X.,Yang, W.,&Xu, B..(2018).Build the Structure of WFSless AO System Through Deep Reinforcement Learning.IEEE PHOTONICS TECHNOLOGY LETTERS,30(23),2033-2036.
MLA Hu, K.,et al."Build the Structure of WFSless AO System Through Deep Reinforcement Learning".IEEE PHOTONICS TECHNOLOGY LETTERS 30.23(2018):2033-2036.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace