Fig. 1. Development of Deep Learning

Fig. 2. ANN and the basic operation of a neuron

Fig. 3. Optical path of using wavefront sensor based on ANN to correct the aberration of two independent mirrors of MMT

Fig. 4. Architecture of wavefront sensor based on ANN

Fig. 5. Result of the PD model of PD-CNN

Fig. 6. Sketch (a) and physical map (b) of the optical system used in the experiment

Fig. 7. Elongation of spots in the SHWFS. Here, r is the distance of the subaperture in the SHWFS from the center (as projected on to the primary mirror), h₀ is the average altitude of the sodium layer, σ_NA is the thickness of the sodium layer and z is the zenith angle

Fig. 8. Experimental result of centroid computation for Shack-Hartmann wavefront sensor in extreme situations based on the SHNN

Fig. 9. Wavefront sensing by using multiple laser guide stars

Fig. 10. CARMEN with MLP

Fig. 11. Example of the topology of CARMEN with CNN

Fig. 12. ISNet architecture

Fig. 13. Statistical results of the RMS wavefront error of five methods in wavefront detection. Each kind of phase screen contains 100 datasets

Fig. 14. Physical process and flow chart of the CS-WFS method