Poyneer Lisa A, Véran Jean-Pierre
Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, California 94550, USA.
J Opt Soc Am A Opt Image Sci Vis. 2010 Nov 1;27(11):A223-34. doi: 10.1364/JOSAA.27.00A223.
In many scenarios, an adaptive optics (AO) control system operates in the presence of temporally non-white noise. We use a Kalman filter with a state space formulation that allows suppression of this colored noise, hence improving residual error over the case where the noise is assumed to be white. We demonstrate the effectiveness of this new filter in the case of the estimated Gemini Planet Imager tip-tilt environment, where there are both common-path and non-common-path vibrations. We discuss how this same framework can also be used to suppress spatial aliasing during predictive wavefront control assuming frozen flow in a low-order AO system without a spatially filtered wavefront sensor, and present experimental measurements from Altair that clearly reveal these aliased components.
在许多情况下，自适应光学（AO）控制系统在存在时间上非白噪声的环境中运行。我们使用具有状态空间公式的卡尔曼滤波器，该滤波器能够抑制这种有色噪声，从而相较于假设噪声为白噪声的情况，降低残余误差。我们在估计的双子座行星成像仪倾斜环境（存在共光路和非共光路振动）的案例中，展示了这种新型滤波器的有效性。我们讨论了在假设低阶AO系统中流动冻结且没有空间滤波波前传感器的情况下，如何使用相同的框架在预测波前控制期间抑制空间混叠，并展示了来自牵牛星的实验测量结果，这些结果清楚地揭示了这些混叠分量。
