Wavefront Distortion Prediction Method Based On Motion Estimation

Adaptive Optics(AO)system can compensate the wavefront distortion by realtime detection and correction,and improve the resolution and image quality of the optical system,aiming at the random disturbance of atmospheric turbulence in time and space.In actual control system,the time delay is about 2 ～ 3 times of system sampling period,which causes the wavefront distortion correction lag error and limits the control bandwidth.For the observation of the faint targets,it is necessary to have a bright enough guide star nearby to provide real-time turbulence wavefront distortion information,but the region with good correction effect is very small.Therefore,combined with the characteristics of atmospheric turbulence and the principle of adaptive optical control system,the prediction and estimation methods of wavefront distortion in time and space domain are studied in this paper.Based on the assumption of atmospheric frozen turbulence,atmospheric turbulence is affected by transverse wind,and the wavefront distortion is predictable in time domain.A wavefront distortion prediction method based on motion estimation in time domain is proposed.The template matching algorithm is used to estimate the transverse wind parameters,and the current images are processed to achieve the prediction,according to the past and present wavefront restoration images.The estimation performance for atmospheric turbulent motion parameters is evaluated by comparing the wavefront sequence images of different wind speed and direction.The method is optimized for the influence of other aberrations in the system,and simulation data with different sampling frequencies are compared to evaluate the applicability of the prediction method.The results show that the error of wind speed estimation is less than 20%,and the error of wind direction estimation is less than 10 deg.Compared with the least recursive square(RLS)model prediction method,the prediction effect is better,especially in the case of obvious variation tendency of wavefront restoration image.Finally,using the actual Sirius observation data,the prediction method is verified,which exhibits its unique advantages.Based on the correlation of multi-target spatial transmission,a spatial wavefront distortion estimation method based on motion estimation is proposed.The template matching algorithm is used to analyze the similarity of multiple targets in space,calculate the maximum overlap,process the wavefront information based on the relative coordinate positions,and estimate the wavefront of other positions.By comparing the data of different target distribution and different target quantity,the estimation effect caused by the change of target relative position is evaluated.The results show that the more symmetrical the target distribution is,the estimation performance is better,and the angular separation remains in the range of 30 arcseconds to 55 arcseconds.Under the same angular separation between targets,the estimation effect of three known targets is better than that of two known targets,and the estimation range is larger.The estimation method has better performance than the direct interpolation method.