Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userEfficient implementation of pseudo open loop control for adaptive optics on Extremely Large Telescopes
88
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Closed-loop adaptive optics systems which use minimum mean square error wavefront reconstruction require the computation of pseudo open loop wavefront slopes. These techniques incorporate a knowledge of atmospheric statistics which must therefore be represented within the wavefront slope measurements. These pseudo open loop slopes are computed from the sum of the measured residual slopes and the reconstructed slopes that would be given if the deformable mirror was flat, generally involving the multiplication of an interaction matrix with actuator demands from the previous time-step. When using dense algebra, this multiplication is computationally expensive for Extremely Large Telescopes, requiring a large memory bandwidth. Here we show that this requirement can be significantly reduced, maintaining mathematical correctness and significantly reducing system complexity. This therefore reduces the cost of these systems and increases robustness and reliability.
rate research
Read More
Advanced adaptive optics (AO) instruments on ground-based telescopes require accurate knowledge of the atmospheric turbulence strength as a function of altitude. This information assists point spread function reconstruction, AO temporal control techniques and is required by wide-field AO systems to optimize the reconstruction of an observed wavefront. The variability of the atmosphere makes it important to have a measure of the optical turbulence profile in real time. This measurement can be performed by fitting an analytically generated covariance matrix to the cross-covariance of Shack-Hartmann wavefront sensor (SHWFS) centroids. In this study we explore the benefits of reducing cross-covariance data points to a covariance map region of interest (ROI). A technique for using the covariance map ROI to measure and compensate for SHWFS misalignments is also introduced. We compare the accuracy of covariance matrix and map ROI optical turbulence profiling using both simulated and on-sky data from CANARY, an AO demonstrator on the 4.2 m William Herschel telescope, La Palma. On-sky CANARY results are compared to contemporaneous profiles from Stereo-SCIDAR - a dedicated high-resolution optical turbulence profiler. It is shown that the covariance map ROI optimizes the accuracy of AO telemetry optical turbulence profiling. In addition, we show that the covariance map ROI reduces the fitting time for an extremely large telescope-scale system by a factor of 72. The software package we developed to collect all of the presented results is now open source.
The performance of a wide-field adaptive optics system depends on input design parameters. Here we investigate the performance of a multi-conjugate adaptive optics system design for the European Extremely Large Telescope, using an end-to-end Monte-Carlo adaptive optics simulation tool, DASP. We consider parameters such as the number of laser guide stars, sodium layer depth, wavefront sensor pixel scale, number of deformable mirrors, mirror conjugation and actuator pitch. We provide potential areas where costs savings can be made, and investigate trade-offs between performance and cost. We conclude that a 6 laser guide star system using 3 DMs seems to be a sweet spot for performance and cost compromise.
The performance of a wide-field adaptive optics system depends on input design parameters. Here we investigate the performance of a multi-object adaptive optics system design for the European Extremely Large Telescope, using an end-to-end Monte-Carlo adaptive optics simulation tool, DASP, with relevance for proposed instruments such as MOSAIC. We consider parameters such as the number of laser guide stars, sodium layer depth, wavefront sensor pixel scale, actuator pitch and natural guide star availability. We provide potential areas where costs savings can be made, and investigate trade-offs between performance and cost, and provide solutions that would enable such an instrument to be built with currently available technology. Our key recommendations include a trade-off for laser guide star wavefront sensor pixel scale of about 0.7 arcseconds per pixel, and a field of view of at least 7 arcseconds, that EMCCD technology should be used for natural guide star wavefront sensors even if reduced frame rate is necessary, and that sky coverage can be improved by a slight reduction in natural guide star sub-aperture count without significantly affecting tomographic performance. We find that adaptive optics correction can be maintained across a wide field of view, up to 7 arcminutes in diameter. We also recommend the use of at least 4 laser guide stars, and include ground-layer and multi-object adaptive optics performance estimates.
Predictive wavefront control is an important and rapidly developing field of adaptive optics (AO). Through the prediction of future wavefront effects, the inherent AO system servo-lag caused by the measurement, computation, and application of the wavefront correction can be significantly mitigated. This lag can impact the final delivered science image, including reduced strehl and contrast, and inhibits our ability to reliably use faint guidestars. We summarize here a novel method for training deep neural networks for predictive control based on an adversarial prior. Unlike previous methods in the literature, which have shown results based on previously generated data or for open-loop systems, we demonstrate our networks performance simulated in closed loop. Our models are able to both reduce effects induced by servo-lag and push the faint end of reliable control with natural guidestars, improving K-band Strehl performance compared to classical methods by over 55% for 16th magnitude guide stars on an 8-meter telescope. We further show that LSTM based approaches may be better suited in high-contrast scenarios where servo-lag error is most pronounced, while traditional feed forward models are better suited for high noise scenarios. Finally, we discuss future strategies for implementing our system in real-time and on astronomical telescope systems.
The highest resolution images ever taken in the visible were obtained by combining Lucky Imaging and low order adaptive optics. This paper describes a new instrument to be deployed on the WHT 4.2m and GTC 10.4 m telescopes on La Palma, with particular emphasis on the optical design and the expected system performance. A new design of low order wavefront sensor using photon counting CCD detectors and multi-plane curvature wavefront sensor will allow dramatically fainter reference stars to be used, allowing virtually full sky coverage with a natural guide star. This paper also describes a significant improvements in the efficiency of Lucky Imaging, important advances in wavefront reconstruction with curvature sensors and the results of simulations and sensitivity limits. With a 2 x 2 array of 1024 x 1024 photon counting EMCCDs, AOLI is likely to be the first of the new class of high sensitivity, near diffraction limited imaging systems giving higher resolution in the visible from the ground than hitherto been possible from space.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
