人眼像差校正线性二次高斯优化控制研究

Linear Quadratic Gaussian Optimal Control for Human Eye Aberration Correction

为获得清晰的高分辨率人眼眼底视网膜图像,自适应光学系统中变形镜必须能够实时跟踪并补偿人眼中随时间变化的像差信息。对波前像差特别是动态像差的校正能力不仅取决于变形镜等硬件的性能,还与自适应光学系统中的控制算法密切相关。在不加大硬件复杂度的基础上,介绍了一种基于Kalman滤波的线性二次高斯(LQG)人眼像差校正最优控制模型。首先分析了自适应光学系统的离散性,证明在离散模式下研究自适应光学系统的可行性;然后建立了基于Kalman滤波的LQG优化控制模型,并给出基于LQG优化控制的像差校正控制算法;最后通过仿真实验验证了基于LQG优化控制的像差校正算法对动态人眼波前像差校正的可行性和有效性。

In order to obtain a clear retinal image resolution of the human eye, the deformable mirror in adaptive optics system must be able to track and compensate the eyes aberration information in real time. The capability of wavefront aberration correction, especially the dynamic wavefront aberration, is not only depending on the performance of the deformation mirror and other hardwares, but also closely related to the control algorithm of adaptive optics system. Without increasing hardware complexity, a human eye aberration correction optimal control model based on Kalman filter and linear quadratic Gaussian （LQG） control is proposed. Firstly, the dispersion of adaptive optics system is analyzed and it is shown that the study of adaptive optics system under the discrete model is feasible. Then, the LQG optimal control model based on the Kalman filtering is established, and the aberration correction algorithm based on LQG optimal control model is proposed. Finally, the simulation experiment demonstrates the feasibility and effectiveness of the proposed algorithm.