Stochastic parallel gradient descent algorithm for adaptive optics system based on Zernike mode

The convergence rate can be the limit of adaptive optics without a wave-front sensor in real-time applications. An improvement on the convergence rate of the stochastic parallel gradient descent(SPGD) algorithm is discussed by combing modal method with zonal method. Based on the SPGD algorithm, adaptive optics systems are simulated with a 61-element deformable mirror. Results show that the modified SPGD algorithm can converge much faster than that based only on zonal method when the same correction effect is obtained.