基于深度学习的无波前探测自适应光学系统研究进展

Research progress in deep learning based WFSless adaptive optics system

近年来自适应光学(AO)系统向着小型化和低成本化趋势发展,无波前探测自适应光学(WFSless AO)系统由于结构简单、应用范围广,成为目前相关领域的研究热点。硬件环境确定后,系统控制算法决定了WFSless AO系统的校正效果和系统收敛速度。新兴的深度学习及人工神经网络为WFSless AO系统控制算法注入了新的活力,进一步推动了WFSless AO系统的理论发展与应用发展。在回顾前期WFSless AO系统控制算法的基础上,全面介绍了近年来卷积神经网络(CNN)、长短期记忆神经网络(LSTM)、深度强化学习在WFSless AO系统控制中的应用,并对WFSless AO系统中各种深度学习模型的特点进行了总结。概述了WFSless AO技术在天文观测、显微成像、眼底成像、激光通信等领域的应用。

In recent years,Adaptive Optics(AO)system is developing towards miniaturization and low cost.Because of its simple structure and wide application range,wavefront sensorless(WFSless)AO system has become a research hotspot in related fields.Under the condition that the hardware environment is determined,the system control algorithm determines the correction effect and convergence speed of WFSless AO system.The emerging deep learning and artificial neural network have injected new vitality into the control algorithms of WFSless AO system,and further promoted the theoretical and practical development of WFSless AO.On the basis of summarizing the previous control algorithms of WFSless AO system,the applications of convolution neural network(CNN),long-term memory neural network(LSTM)and deep reinforcement learning in WFSless AO system control in recent years are comprehensively introduced,and characteristics of various deep learning models in WFSless AO system are summarized.Applications of WFSless AO system in astronomical observation,microscopy,ophthalmoscopy,laser telecommunication and other fields are outlined.