Adaptive phase aberration correction based on imperialist competitive algorithm

Abstract

We investigate numerically the feasibility of phase aberration correction in a wavefront sensorless adaptive optical system, based on the imperialist competitive algorithm (ICA). Considering a 61-element deformable mirror (DM) and the Strehl ratio as the cost function of ICA, this algorithm is employed to search the optimum surface profile of DM for correcting the phase aberrations in a solid-state laser system. The correction results show that ICA is a powerful correction algorithm for static or slowly changing phase aberrations in optical systems, such as solid-state lasers. The correction capability and the convergence speed of this algorithm are compared with those of the genetic algorithm (GA) and stochastic parallel gradient descent (SPGD) algorithm. The results indicate that these algorithms have almost the same correction capability. Also, ICA and GA are almost the same in convergence speed and SPGD is the fastest of these algorithms.

An Automatic Optimized Method for a Digital Optical Phase Conjugation System in Focusing through Scattering Media

“…In order to obtain the optimal compensation effect and improve the quality of DOPC, it is necessary to seek an efficient and accurate algorithm to acquire the optimal Zernike coefficients. Among the optimization algorithms, the multipopulation genetic algorithm (MPGA) [36][37][38] can be the proper algorithm to combine with DOPC for its intrinsic advantages, including the small amount of computation, strong convergence, and excellent real-time performance, especially in solving global optimization problems. Furthermore, the fitness function of the algorithm is crucial to the optimization result by MPGA.…”

Reconstruction of two interfering wavefronts using Zernike polynomials and stochastic parallel gradient descent algorithm

“…For this purpose, different optimization algorithms, such as genetic, simulated annealing, imperialist competitive [11], and stochastic parallel gradient descent (SPGD) [12], can be used. We apply the SPGD algorithm, since the effectiveness of its search capability and its high convergence speed have already been demonstrated, and so this algorithm has been widely used for aberration correction even in real time [13].…”