Wavefront correction of model-based sensorless adaptive optics system

Abstract

Compared with the conventional adaptive optics (AO) system, the sensorless AO system can greatly reduce the complexity of system. The slow convergence speed is the biggest defect of this type of system because the existing sensorless adaptive optics systems mostly use various blind optimization algorithms as the control algorithm of the system. This paper presents a closed-loop control algorithm based on model for the sensorless AO system. An adaptive optics system simulation platform is established with a 61-element deformable mirror and a CCD imaging device. The convergence speed and the correction capability are investigated through correcting wavefront aberrations, which conform to Kolmogrov spectrum and are composed of 102 orders Zernike mode. Results are compared to those of SPGD (Stochastic parallel gradient descent), the most commonly used control algorithm for sensoress AO systems. Research results show that both algorithms can obtain the correction capability getting close to the best effect of 61-element deformable mirror. The model-based sensorless AO system requires 103 times measurement of the far field spot, but the SPGD-based sensorless AO system needs about 1500 times. Therefore, the control algorithm provided can greatly improve the convergence speed of sensorless AO systems.