The experimental adaptive optics system, currently developed at the University of Hawaii, is now equipped with a VME-based control system, and a high sensitivity wave-front sensor. The sensor uses an array of 13 photon-counting avalanche photodiodes which enable the system to work with faint reference or `guide' sources, as faint as magnitude 15. Results of the first successful observing runs are described here.

This paper reports progress in the use of laser beacons for compensating the images of faint objects using adaptive optics. The system describe is located at the USAF Phillips Laboratory's Starfire Optical Range near Albuquerque, New Mexico on a 1.5 m telescope. The adaptive optics system uses a 241 actuator continuous facesheet deformable mirror. A copper vapor laser beacon focused at 10 km range is used to sense higher-order distortions. A natural guide star is used for sensing full aperture tilt. The limiting factor for long exposure performance of the system is tracking--the correction of full aperture tilt. The main goal of the work reported here was to determine the performance of the tracking system under conditions of weak signal but good higher order compensation (similar to conditions encountered in laser beacon adaptive optics). Single axis rms track errors of approximately 20 milliarcsec were achieved for track signal levels equivalent to stars of 13^th magnitude. However, we cannot track stars fainter than 9 - 10^th magnitude because the laser induces very low level, long lived phosphorescence in the optical train. This appears to be a fundamental limitation for our system as it is presently configured.

The status and plans for a multi-phase program to build adaptive optics (AO) user facilities for one of the Keck telescopes is presented. The planned facilities include (1) fast tip/tilt correction, (2) near infrared AO with natural stars, and potentially (3) a near infrared AO facility with a single laser beacon. Description of these facilities and their implementation on the telescope are described. In addition, descriptions of the current and future suite of scientific instruments that would take advantage of adaptive optics are provided. Problems and concerns associated with implementing adaptive optics facilities at Keck (e.g., a segmented primary, a 10 meter baseline, rotation of a non-symmetric pupil, etc.) are discussed.

Adaptive optics is one of the main features of the Very Large Telescope of the European Southern Observatory (ESO). Use of the large telescopes for high spatial resolution imaging, high spectral resolution spectroscopy and astronomical spatial interferometry depends substantially on the availability of adaptive optics, which allows diffraction limited imaging for each individual telescope in the near infrared wavelength range (2.2 to 5 microns) and partial correction of atmospheric distortions towards the visible wavelength range. Recent decisions from ESO Council to postpone the VLT interferometer and the associated adaptive optics and Coude beams has led us to study the implementation of this capability at the Nasmyth focus. We present here the first preliminary conceptual design which follows the outline of the Coude configuration from the performance point of view. Some scientific pre-studies, such as sky coverage evaluation and statistics of observable objects are summarized. Finally an analysis of the different sources of image degradation after the adaptive optics correction, are described.

Since approximately 5 years, several experimental Adaptive Optics Systems have been tested on different astronomical sites. This paper describes a general purpose AOS designed for new or existing telescopes for diameters ranging from 2 to 4 meters. Investigations on actually available Adaptive Mirrors and Wavefront sensors are made. The use of a bimorph mirror and simplified Shack Hartmann is proposed. Results of simulation to evaluate the Strehl ratios for different wavelengths are given versus seeing conditions and star magnitudes. This leads to the conclusion that low cost, turn key AOS are a new attractive possibility for Astronomy.

Wavefront correctors used in astronomical telescopes today are usually zonally controllable devices such as deformable or segmented mirrors, having a large number of independently adjustable actuators. In this paper we describe an entirely different approach, using intrinsically rigid optical components such as small lenses or mirrors, which are displaced and tilted to implement the wavefront correction, each element spanning the aperture at a reduced beam diameter. Modal corrections approximating Zernike polynomials can be produced with five degrees of freedom: linear displacements along the x, y and z axes, plus rotation about the x and y axes. This approach provides a simple and efficient method for low-order wavefront compensation. In some cases the active elements may replace existing optical components.

A serious impediment to the development of large aperture optics systems for correction of atmospheric turbulence is the very large number of wavefront correction subsystems (each consisting of an actuator, an actuator drive circuit, and a wavefront measurement element) which are required, and the attendant cost of these subsystems. If wavefront correction is to be applied to a fully-filled circular pupil several meters in diameter, several thousand of these subsystems may be needed. This situation could, however, be ameliorated if the pupil were not fully-filled. Here, we demonstrate that the diffraction-limited resolution of a full circular pupil can be approached or equalled with a synthetic aperture optical system consisting of an elongated pupil which rotates. Separate exposures of the object scene are made at each of several rotational positions of the pupil, and the final scene is reconstructed for these exposures. No optical phase information is needed for the image reconstruction. The polished area corresponding to the elongated pupil could be an order of magnitude smaller than that of the full pupil, and thus the number of adaptive optics subsystems could be decreased by a similar factor. The technique does not allow us to reach the same limiting magnitude as does the full, diffraction-limited pupil. However, on somewhat brighter objects, it could achieve the angular resolution of the full pupil without the complication and cost of adaptively correcting the entire pupil.

The central core intensity of a stellar image observed by a ground-based telescope can be maximized by a judicious balancing of the adaptive optics system and the size of the telescope entrance aperture. For a given aperture, increasing the number of degrees of adaptive optics turbulence compensation will maximize the brightness of the central core. However, for an observatory using an adaptive optics system with a fixed number of degrees-of-freedom, the largest aperture available will not necessarily result in a maximized image central core. The negative effects of atmospheric turbulence, roughly proportional to (formula available in paper) cannot always be compensated by the increased light gathering ability of a larger aperture (proportional to D²). It is shown and verified through simulation that the optimum aperture diameter is a function of N^p r_o where N is the number of adaptive optics degrees of freedom and r_o is the seeing cell size. The simulations show that the exponent p is related to the control algorithm or, more precisely, the figure-of-merit used to drive the deformable mirror actuators. Optimizing the useful aperture of the telescope/adaptive optics system is a strategy that can make use of the variation in site seeing conditions and benefit the astronomer by increasing the available number of observable science objects or reducing the observing time.

Wavefront correctors acting in a single plane can only correct for phase distortion on a beam from a single direction. If the distortion arises in a distributed medium beams from other directions will be at best partially corrected. The region over which good correction occurs can be extended by using multiple correctors imaged at different locations in the medium. A procedure is derived here for locating the correctors to maximize the residual anisoplanatic error and this minimum error is evaluated for several atmospheric models. This procedure also applies to the simulation of distributed distortion by discrete phase screens.

We describe an analytic model for the point spread function in astronomical imaging through atmospheric turbulence with modal adaptive correction. The model includes the effects of seeing quality, degree of modal correction, angular anisoplanaticism, and wavefront sensor noise. Anisoplanatic effects are contrasted for cases where the correcting mirror is conjugate with the telescope pupil and with the atmospheric turbulence. The properties of the point spread function are described in terms of its shape, Strehl ratio, and included energy widths.

In this paper we discuss the performance of low order adaptive optics (AO) systems. We present improved calculations of Strehl ratio achievable at various D/r_o ratios for AO systems up to 7^th order. Additionally we present calculations concerning the throughput of low order AO systems used in conjunction with spectrometers. We then show the results of a detailed simulation of the expected interaction of the University of Hawaii prototype AO system with spectrometers through the V,I,J,H,K optical bands. This simulation shows that AO spectroscopy performance should always be acceptable in the near IR (J,H,K), and often in the visible (V,I). We show the results of our initial efforts at deconvolving the AO PSF. These results indicate that deconvolution is likely to be quite widely applicable to AO images. Finally we discuss how it might be possible to extend the applicability and accuracy of deconvolution using the AO drive and wavefront error signals. The following results are for on-axis system, the effects of atmospheric anisoplanitism have not been addressed in this paper.

The effects of turbulence-induced anisoplanatism cause the performance of an adaptive optic system to be dependent on the angular separation between the object wavefront being corrected and the wavefront of the reference source. One method of quantifying this angular dependent performance is through the average optical transfer function (OTF). An equally important measure is the variance of the OTF. The variance is used together with the average OTF to define a signal-to-noise ratio (SNR). Knowledge of the SNR is essential in any image restoration scheme to determine the spatial frequency limit of the restoration process. This paper presents the diffraction based theory and a method of computing the object angle dependent average OTF and corresponding SNR of an adaptive-optics system. These quantities are computed using normalized correction functions that are valid for a wide range of atmospheric turbulence profiles and angular correction geometries. This method allows for rapid calculations under changing atmospheric and geometric conditions.

We identify stellar wave fronts that passed through turbulence as fractals from the Fractional Brownian motion family. Fractal behavior manifests itself both in space and in time. We propose predicting wave fronts, based on earlier measurements at each pixel and at neighboring pixels. If the wind direction is known, mainly upward measurements are needed. The accuracy of the method is improved if atmospheric layers can be measured separately.

We discuss issues in optimizing the design of adaptive optics and laser guide star systems for the Keck Telescope. The initial tip-tilt system will use Keck's chopping secondary mirror. We describe design constraints, choice of detector, and expected performance of this tip-tilt system as well as its sky coverage. The adaptive optics system is being optimized for wavelengths of 1 - 2.2 micrometers . We are studying adaptive optics concepts which use a wavefront sensor with varying numbers of subapertures, so as to respond to changing turbulence conditions. The goal is to be able to `gang together' groups of deformable mirror subapertures under software control, when conditions call for larger subapertures. We present performance predictions as a function of sky coverage and the number of deformable mirror degrees of freedom. We analyze the predicted brightness of several candidate laser guide star systems, as a function of laser power and pulse format. These predictions are used to examine the resulting Strehl as a function of observing wavelength. We discuss laser waste heat and thermal management issues, and conclude with an overview of instruments under design to take advantage of the Keck adaptive optics system.

Over the last several years the practical implementation of adaptive optics to compensate for the atmospheric distortions in large telescopes has become a reality. Of the elements that must be considered in the design of an adaptive optics system, the expected atmospheric turbulence is one of the most important. The usual method for estimating these criteria is to use standard atmospheric models or site specific adaptations of these models. An implicit assumption in these models is that the atmosphere can be treated as an isotropic mass and that the index of refraction variations follow Kolmogorov theory. An analysis of two of these features, smoothed data vice a set of individual turbulence profiles and the influence of a partially non-Kolmogorov atmosphere, was performed for a large adaptive optics system. The results show that performance expectations can vary significantly. Smoothed data tends to over estimate atmospheric effects up to 50%. Non-Kolmogorov effects are less significant introducing differences on the order of 10% for zenith observations. The conclusion is that the designer must pay careful attention to the atmospheric model and the method in which it is employed. The use of multiple phase screens created directly from sonde data are recommended.

We present the design and implementation of a very compact adaptive optics system that senses the return light from a sodium guide-star and controls a deformable mirror and a pointing mirror to compensate atmospheric perturbations in the wavefront. The deformable mirror has 19 electrostrictive actuators and triangular subapertures. The wavefront sensor is a Hartmann sensor with lenslets on triangular centers. The high-bandwidth steering mirror assembly incorporates an analog controller that samples the tilt with an avalanche photodiode quad cell. An f/25 imaging leg focuses the light into a science camera that can either obtain long-exposure images or speckle data. In laboratory tests overall Strehl ratios were improved by a factor of 3 when a mylar sheet was used as an aberrator. The crossover frequency at unity gain is 30 Hz.

This article reviews the relation between spatial-temporal characteristic of atmospheric perturbation and several parameters of adaptive optics system. The altitude profile models of atmospheric turbulence intensity and turbulent outer scale, as well as wind velocity, are presented. Based on these models the transverse coherence length, the isoplanatic angle, and the critical time constant are calculated. Some properties of which are indicated. The internal errors of an adaptive optics system are discussed. The typical value of each error is presented. Lastly, the Strehl ratio of an adaptive optics system is considered.

We derive an expression for the wavelength (lambda) _o giving maximum resolution for an adaptive-optics compensated telescope. An approximate expression for average on-axis intensity is written to account for the competing effects of diffraction and residual (post-compression) phase error; this expression is then differentiated with respect to the imaging wavelength (lambda) to yield (lambda) _o. The analytically predicted (lambda) _o is compared to simulation results and correspondence is shown to be good at widely separated seeing conditions and adaptive optics geometries.

During 1993 we have upgraded our adaptive optics analysis software package to execute under an interactive, graphical user interface. This software may be used to evaluate a wide range of adaptive optics concepts, including constellations of multiple guide stars, multiconjugate deformable mirror configurations, and integrated adaptive optics/image postprocessing techniques. Adaptive optics system performance is evaluated in terms of mean optical transfer functions, mean point spread functions, and related quantities. These estimates account for the integrated effects of error sources including servo lag, fitting error, sensor noise, and anisoplanatism. They are computed using minimal variance wavefront reconstruction algorithms which will optimize system performance for the given operating conditions.

This study examines the impact of optical aberration and detector spatial quantization on the detection of nonsolar earth-like planets. It describes the use of a Lyot coronagraph/CCD based system implemented on a space-based segmented telescope platform. The coronagraph consists of a cross shaped apodizer in the first focal plane followed by a 75% transmission circular corner Lyot stop at the pupil plane. At the final focus plane, a 2D CCD focal plane array is introduced. The pixel elements on the CCD are varied in size to examine the degrading aspects of finite spatial quantization. Performance results are described in terms of three detectability metrics evaluated for various star-to-planet separations. Robustness of these metrics to segmentation phasing and tilt errors in the primary mirror are discussed. The effect of planet placement within a pixel on signal variability and the utility of detectability metrics is also examined.

Adaptive optics has been studied for several years as a method to overcome the effects of atmospheric turbulence on ground based astronomy. It is often assumed that full order correction, with one or more actuators per r0, is necessary to achieve any performance improvement in imaging quality with the AO system. Recently it has been shown, both with theoretical analysis and experimental measurements, that even low order adaptive optics (LOAO) can provide image improvements especially when used in conjunction with some sort of computer post processing. In this paper we consider some performance metrics for use in determining expected results from a LOAO system when used for long exposure and short exposure (speckle) imaging. The results of computer simulations are compared to data from a recent experiment on the 1.5 meter telescope at the Starfire Optical Range.

We present an analysis of the tracking error for a laser guide star adaptive optics system with a quadrant detector using the astronomical science target as the reference source. A simple formula relating the tracking error to the SNR and the image profile in the detector plane allows the tracking error for any object and system point spread function to be computed. These calculations show that many types of faint (20 magnitude) science target objects can be used to track provided a high Strehl is achieved at 2.3 micron. This eliminates the need for an off-axis guide star. As an example of a resolved target source, we calculate the tracking error for the spiral galaxy M31 placed at distances out to 100 Mpc. For a given target object spectral energy distribution, an optimal wavelength is obtained from a signal-to-noise calculation. For a given tracking wavelength, the limiting flux is determined by the adaptive optics system's Strehl ratio at the tracking wavelength.

Artificial beacons generated at the sodium layer have many advantages for adaptive optics (AO) systems in astronomy. The source can be sufficiently bright to achieve high Strehl ratios at infrared wavelengths and because the target object can often be used as the tracking source, photometric field errors can be much reduced compared to off-axis natural beacons. The problem has been to develop a laser that could be tuned to the sodium D₂ line and focused into a compact spot without saturating the layer. Both CW and sum frequency lasers are well suited for this work will be used for the ChAOS AO system described elsewhere in this conference. In this paper we discuss some of the parameters that determine how well CW work, the limitations of pulsed lasers and the development of a 10 watt diode laser pumped sum frequency laser for use in astronomy.

We have obtained the first measurements with a sodium laser beacon of focus anisoplanatism over large aperture, at the Multiple Mirror Telescope. In complementary studies, the atmospheric turbulence at the high altitude and on the large scale responsible for the measured focus anisoplanatism was explored by observations of binary stars of different separations. We confirm the predictions of Kolmogorov theory, and derive an effective height for the turbulence of 5050 m above the telescope. These results confirm that the sodium laser guide star planned for use with the 6.5 m telescope conversion of the MMT in 1996 will allow diffraction limited infrared observations in the H and K bands.

This paper discusses the design and analysis of laser-guided adaptive optic systems for the large, 8 - 10 meter class telescopes. We describe a technique for calculating the expected modulation transfer function and the point spread function for a closed loop adaptive optics system, parameterized by the degree of correction and the seeing conditions. The results agree closely with simulations and experimental data, and validate well known scaling law models even at low order correction. Scaling law model analysis of a proposed adaptive optics system at the Keck telescope leads to the conclusion that a single laser guide star beacon will be adequate for diffraction limited imaging at wavelengths between 1 and 3 micrometers with reasonable coverage of the sky. Cone anisoplanatism will dominate wavefront correction error at the visible wavelengths unless multiple laser guide stars are used.

The measuring principle and basic composition of the CLIAM (Coherence Length and Isoplanatic Angle Measurer) are presented and the results measured at Yunnan Observatory in Kunming are reported in this paper. The measured results are compared with the calculated values based on C²_n altitude profile measured by using ballon-borne macrothermal probe. Finally, some factors, which could have an effect on measuring validity, are analyzed.

We have installed a high-speed camera system at the Keck Telescope, to be used for studying and monitoring atmospheric seeing as well as for telescope diagnostic purposes. This instrument, which consists of a Dalsa camera with a 64 X 64 pixel CCD, a 4 Megabyte Epix frame grabber, and a 486 computer, records sequences of 1248 frames at 181 Hz and 0.2 arcsecond resolution. We note that the Keck Telescope, by virtue of its 10 meter baseline as well as its ability to separate images formed by any or all of its 36 primary mirror segments, is ideally suited to seeing studies, in particular to those involving relatively long baselines and aperture-aperture correlations of wavefront aberrations. We present power spectra for atmospheric wavefront tilts for the primary mirror segments. In general they show the power law frequency dependance expected on theoretical grounds. However the measured segment-to-segment correlations are systematically smaller than theory predicts by a significant factor. It is possible that this effect is a manifestation of a finite outer scale of turbulence.

The architecture and major system components of the sodium-layer laser guide star system at LLNL will be described, and experimental results reported. The subsystems include the laser system, the beam delivery system including a pulse stretcher and beam pointing control, the beam director, and the telescope with its adaptive-optics package.

Adaptive correction of infrared images formed by large telescopes will be possible with the aid of a single laser guide star projected along the telescope axis. The fundamental limit set by focus anisoplanatism of a sodium beacon allows correction to the diffraction limit in the 1.65 and 2.2 micron bands under typical conditions at a good site. A 1 arcsec `star' of V magnitude approximately equals 9 is then required to reduce photon noise to negligible level. In this paper we report on recent tests of continuous wave dye lasers at wavelength 589 nm, used to create sodium beacons above the Multiple Mirror Telescope. A coherent ring dye laser operating at 2.5 W average power and projected as circularly polarized light yielded a beacon of m_v equals 9.85, and a Livermore-designed standing wave laser operating at 1.7 W and projected as linearly polarized light yielded m_v equals 10.4. Beacon sizes close to 1 arcsec were achieved. The 9th magnitude goal should be achieved by using more advanced cw dye lasers with 20% conversion efficiency of a 25 W argon ion pump laser, and from improvements in beam projection efficiency. We describe a refractive beam expander under construction for the 6.5 m conversion of the MMT.

The design and background data for a sodium layer laser guide star system to be installed on the 3 meter telescope at Lick Observatory is presented. A 30 W dye laser at 589 nm and 10 kHz will be mounted on the telescope and will be pumped by fiber coupled, frequency doubled YAG laser located in a separate room.

A sodium guide star has been used to sense and correct atmospheric aberration during two runs at the Multiple Mirror Telescope (MMT). For the first run in 1993 May, the artificial star was created by a 0.5 W beam from a continuous- wave dye laser tuned to the D₂ resonance line, projected from a telescope centered and coaxial with the main array of six 1.8 m mirrors. Scattering by the mesospheric sodium layer produced an artificial beacon equivalent in brightness to a natural star of visual magnitude 12.5, and of angular extent 1'.2 full width at half maximum (FWHM). During the second run in 1994 February, a 1.7 W dye laser was used to generate an artificial guide star of visual magnitude 10.4, and 1'.1 FWHM. In each case, the beacon was used by the MMT adaptive optics system to compensate in real time for atmospherically- induced differential image motion between the six mirror elements, at correction rates of up to 76 Hz. In the latter experiment, global wavefront tilt correction using a natural reference star was added, giving complete adaptive control. Simultaneously recorded images of a natural star coincident with the laser beacon show significantly reduced width and an increase in Strehl ratio of almost a factor of two.

In a program with the Office of Naval Research and the Innovative Science and Technology directorate of BMDO, we have designed, built and tested a large adaptive optical system for the correction of a visible wavelength laser beam. A visible artificial laser guide star (frequency-doubled Nd:YAG laser with wavelength of 0.532 micrometers ) is used as the source for the reference wavefront. A shearing interferometer which uses a narrow optical bandwidth and has 500 subapertures is employed to sense wavefront distortion. The sensor is used to control a 500-segment deformable mirror which is integrated with a 1 m telescope. In this paper, we will present results for the first known compensated laser illumination tests of an uncooperative low Earth orbit (LEO) satellite using an active point ahead mirror. The results of these tests using a first generation low-power laser beacon at D/r_o equals 25 show improvement of over a factor of 5 in peak energy at the diffraction limited resolution of 0.1 arc sec for stellar images. The amount of light reflected from a large, diffuse LEO satellite was improved by a factor of approximately 2 using the adaptive optical system. In the Spring of 1994, a custom, high powered illuminator laser will be implemented to facilitate novel active tracking and imaging applications.

The first astronomical results using a seven segment edge-matched adaptive mirror are presented. The mirror is a prototype for adaptive secondary mirrors, consisting of seven flat hexagonal mirrors, with inductive edge sensors to measure piston errors, and piezoelectric actuators. A Hartmann- Shack wavefront sensor uses the visible light for centroiding information while the IR image is formed on a 58 X 62 InSb array at 2.2 microns. In addition, a number of mathematical models are presented that illustrate how well the system should perform when correcting atmospheric turbulence.

For the past three years, Steward Observatory's adaptive optics program has been directed toward developing and testing techniques required to correct large telescopes to the diffraction limit. Previously, we have reported on a specific design for an 8 m sodium-laser adaptive system capable of achieving a strong diffraction-limited core at imaging wavelengths of 1.6 - 2.2 microns, with very good sky coverage for image motion sensing using infrared field stars. Recent experiments at the 6.9 m Multiple Mirror Telescope (MMT) have provided the first large aperture measurements of off-axis anisoplanatism for natural stars and focus anisoplanatism for an artificial sodium beacon. This paper summarizes predicted performance for adaptive optics planned for the upgraded MMT, after conversion in 1996 to a single-mirror telescope using the 6.5 m primary mirror developed at Steward's Mirror Laboratory. Performance estimates are obtained using the real MMT atmospheric data and lead to component specifications which are realizable in the near term, permitting the system to be developed and integrated for operation soon after the telescope upgrade is completed.

Binary adaptive optics involves using a simple two-state algorithm for wavefront correction. The technique lends itself to implementation using ferroelectric liquid crystal spatial light modulators (FLC-SLMs) as half-wave and quarter-wave phase shifters. In this paper we demonstrate further the effect of half-wave and quarter-wave correction, we compare the ideal performance with conventional analogue adaptive optics and we discuss the relevant optical properties of FLC-SLMs.

This paper summarizes adaptive optics performance estimates obtained for the Gemini 8-m telescopes, with an emphasis on relatively low order systems employing Shack-Hartmann wavefront sensors, continuous facesheet deformable mirrors with zonal actuators, and either a natural guide star or a laser guide star in the mesospheric sodium layer. Adaptive optics performance is characterized in terms of the fractional encircled energy within a 0.1 arc sec detector at 2.2 microns under average seeing conditions at Mauna Kea. The increased sky coverage achieved using laser guide stars according to this definition of performance is significant, even for a relatively low order adaptive optics system designed for turbulence compensation at 2.2 microns at a good astronomical site.

An astronomical AO system for use on a 4 m class telescope at visible wavelengths is described. The design of this instrument is based on the University of Durham semi common-user partial-AO system MARTINI. A brief technical resume and recent astronomical results are provided. The MARTINI system was designed on a partial, but non-modal, philosophy which is able to deliver modest image improvements normally associated with low order corrections at small D/r_o. The new system will remove the limited aperture coverage of the MARTINI system and extend its operating philosophy to produce a corrected optical transfer function optimized to scientific goals. The requirements and resulting system design are described. The system is also designed to be accommodated within the general scheme of the UK national AO program which was initiated in 1993. A brief discussion of the design issues involved is provided.

A tip-tilt secondary system has been developed on a 2.3 m telescope. The system, called FASTTRAC, stabilizes image motion to less than 0.1' rms at closed loop corrective frequencies of <EQ 100 Hz. Resolutions of 0.19' and Strehl ratios of 0.14 have been obtained in long exposure images at 1.6 micrometers in seeing conditions of D/r_o approximately 4. FASTTRAC is unique in its ability to use infrared guide stars (K <EQ 8). Recently FASTTRAC was upgraded to accommodate a low read noise CCD, allowing faint visible guide stars to be utilized. We anticipate that guide stars of R equals 17 will be adequate at correction frequencies of 30 Hz.

This paper describes some of the history and the current design of the Chicago Adaptive Optics System (ChAOS). ChAOS is a facility AO system using a sodium layer laser beacon which is due to be permanently mounted on a 3.5 meter ARC telescope located at Apache Point New Mexico. First light is expected in 1995.

Preliminary results are presented from Rayleigh beacon closed loop images of the large asteroids Ceres, Pallas, and Vesta obtained with 1.5 m telescopes of the USAF Phillips Laboratory's Starfire Optical Range. Measurements of the apparent diameters of two post-processed images of Ceres lead to excellent true diameter, and of two images of Pallas to good true values. Measurements of images of Vesta on two occasions lead to the full triaxial ellipsoid solution and rotational pole direction, all in excellent agreement with previous speckle interferometric results. A preliminary iterative blind deconvolution image of Vesta obtained from three 20 sec integrations is presented.

We will describe an adaptive optics system developed for the 1 meter Nickel and 3 meter Shane telescopes at Lick Observatory. Observing wavelengths will be in the visible for the 1 meter telescope and in the near IR on the 3 meter. The adaptive optics system design is based on a 69 actuator continuous surface deformable mirror and a Hartmann wavefront sensor equipped with an intensified CCD framing camera. The system has been tested at the Cassegrain focus of the 1 meter telescope where the subaperture size is 12.5 cm. The wavefront control calculations are performed on a four processor single board computer controlled by a Unix-based system. We will describe the optical system and give details of the wavefront control system design. We will present predictions of the system performance and initial test results.

An artificial neural network has been developed for wavefront reconstruction by processing the centroid displacements of the subaperture images of a 4 X 4 Hartmann sensor. Training was accomplished by computing the first differentials of random but known mixes of aberration polynomials, presenting these at the network input and adjusting the interconnection weights to minimize the Zernike coefficient output errors. After training, the residual rms wavefront errors for noise-free systems were typically within 2%. Similar results were obtained with a 5 X 5 system. As simulated noise was added, the errors increased slowly but were similar in magnitude to those obtained analytically provided that the network had sufficient complexity. At very low SNR values the neural network outperformed the analytic method provided that it had been trained for the same noise level.

During the past 20 years of Adaptive Optics system design, two forms of wavefront sensors-- the Shearing Interferometer (SI) and the Hartmann Sensor (HS)--have predominated. Each of these sensors has advantages and disadvantages which must be considered in the light of the requirements of the particular application. As part of Itek's continuing R&D in adaptive optics, we have developed a new form of sensor which can be viewed as a hybridization of the SI and HS. This sensor is called the Phase Visualization (PV) sensor. The PV sensor uses two orthogonal filters whose reflection and transmission vary linearly with position. The four output images from these filters have intensities which are linearly related to the phase gradient in the incident field. Processing the images results in two gradient maps that can be used in standard reconstructor algorithms.

A photon counting wavefront curvature sensor (WFS) with 13 subapertures suitable for adaptive optics in astronomy has been developed at the University of Hawaii. This sensor is capable of using very faint point sources or slightly extended sources to derive the wavefront signal. The sensitivity of this sensor is continuously variable and can be adjusted in real time to match the seeing conditions at the time. The wavefront sampling geometry has been optimized for correcting the standard atmosphere up to 9 orders expressed in terms of Zernike's. Its output is used in conjunction with a newly developed deformable bimorph mirror for high efficiency correction capabilities. This WFS has successfully been used recently at the CFHT and UKIRT facilities on Mauna Kea on a variety of astronomical objects. Point sources, double stars, planetary nebula, galactic nuclei, and some of the moons of Jupiter have all been successfully attempted. Limiting magnitude has not been explored in great detail at the telescope, but we have taken the system down to magnitude R equals 13.7 (V equals 15) with a 3.6 meter aperture with success. This was achieved during bright time or whilst the full moon was present.

The resolution of optical imaging systems is severely degraded from the diffraction limit by the random effects of the atmosphere. Techniques exist to compensate for the atmospheric turbulence, one of which is adaptive optics. A critical component in the adaptive optics system is the wavefront sensor. Presently, two types of sensors are being used--the Hartmann-Shack Wavefront Sensor and the Shearing Interferometer. Previous studies have compared these two sensors and found them to perform identically for point source beacons. However, to date, no comparison has been performed for an extended beacon and subaperture spacing larger than the correction length of the atmosphere, r_o. This paper examines this problem and compares these two sensors for the above conditions. Results indicate that the sensors have comparable performance when an infinite number of photons are available. However, the photon limited cases indicate superior performance by the Hartmann Sensor.

We present results of simulations involving a curvature-based wavefront sensor which uses an extended pattern as a reference source. The proposed sensor provides measurements of both symmetric and asymmetric aberration terms by comparing the Fourier transforms of two oppositely defocused images. Symmetric terms such as defocus and astigmatism can be measured without regard to the object distribution. The asymmetric terms, such as tip and tilt, rely on averaging the signal over many atmospheric realizations in order to determine the object phase, or on defining an arbitrary reference phase. Only after removal of the object Fourier transform phase can the asymmetric terms be identified. Although this paper reports on preliminary results, we believe the proposed sensor will be useful for both real-time compensation of atmospheric distortions while imaging the Sun, and post-facto compensation of optical misalignments in Earth-pointing satellites.

F. Roddier has proposed a curvature sensing and modal compensation adaptive optics system, currently under development at the University of Hawaii. Several astronomical groups have shown interest in such a system, including the Gemini 8-m Telescope Project, and wonder how it compares with previously demonstrated slope sensing/zonal correction systems. To begin to address this question, we present in this paper an optimal estimator analysis of curvature sensing, based on the method Wallner developed for slope sensing systems.

Atmospheric disturbed wavefront with Kolmogoroff spectrum is represented by the Zernike polynomials suggested by R. Noll in 1976. From the irradiance distribution in two defocal planes, one is in the front of focal plane and the other is in the behind focal plane, the response elements for each term of Zernike polynomials with a given detector array and successively the whole response matrix R have been obtained. For an incoming atmospheric disturbed random wavefront W_R, with its irradiance distribution detected and the matrix R, it is easy to solve the corresponding coefficients for each term of Zernike polynomials and reconstruct the detected wavefront.

A single defocused star image contains sufficient information to uniquely determine the spatial phase fluctuations of the incident wavefront. A sensor which responds to the intensity distribution in the image produces signals proportional to the wavefront curvature within the pupil and the radial slope at the pupil boundary. Unlike Roddier's differential curvature sensing technique, a single-image sensor does not cancel intensity fluctuations due to atmospheric scintillation. However, it has been shown that at typical astronomical sites the scintillation signal is negligibly small. A single-image curvature sensor can theoretically achieve a signal-to-noise ratio of order Q approximately equals r²₀/(lambda) z₀ where r₀ is Fried's correlation length, (lambda) is the wavelength, and z₀ is the root-mean-square distance through the atmosphere, weighted by the refractive index structure constant C²_n. This is more than adequate for AO systems whenever D/r₀ <EQ Q^6/5. Such a sensor can be very simple, optically and mechanically, and has lower detector read noise than a comparable differential system. The concept has been tested in the laboratory by introducing, and detecting, spherical aberration in a simple optical system.

This paper reports the results of some experiments on wavefront sensing at La Palma, Canary Is., using a Shack-Hartmann sensor. From the Shack-Hartmann data, we derive estimates of the seeing parameter r₀ and compare different methods of wavefront reconstruction: a new method of wave front reconstruction based on Bayes' theorem is presented. The design of a Shack-Hartmann sensor for operation at 2 micrometers on UKIRT is described.

CCDs having circular rather than rectilinear symmetry offer some advantages for low-order wavefront sensing. A series of imagers having 24 sectors and 12 radial zones (plus frame store) has been developed. Three different output structures have been utilized in order to investigate low-noise operational limits.

The Adaptive Optics CCD (ACCD) is a 32 by 64 frame transfer device offering low noise performance at high frame rates using skipper amplifier technology to achieve low noise performance. An array of 32 parallel readouts provides for fast framing at 1.5 KHz with a design goal of 1.5 electrons (rms) noise. Potential applications for the ACCD are in wavefront sensing, tip/tilt sensing, and fast tracking. Eleven devices has been successfully packed and tested. An 8 electron rms noise floor has been measured using a single read from the skipper amplifier. Multiple reads are expected to diminish the noise to 2.5 electrons rms. Design goals are expected to be reached by changing the spacing between output amplifiers, using the newly tested low noise CUBIC amplifier, and through thinning and backside illumination.

A series of small rapid-framing CCD designs have been designed, fabricated, and tested for use as adaptive optics sensors for laser guide star image correction. To reduce video bandwidth and thus minimize readout noise, these CCD's employ 16 output amplifiers. Designs with 2 different formats (64 X 64 and 128 X 128) and with three different output amplifiers structures have been produced.

The COAST group in the University of Cambridge have a lot of operational experience of using fast guiding systems to control the pointing of up to four separate telescopes simultaneously as well as a lot of operational experience of using array CCD systems and avalanche photodiode photon counting systems for signal detection. This paper looks at the advantages and disadvantages of these systems for fast guiding and wavefront sensing applications.

This paper presents an adaptive optics (AO) system for a large IR telescope. The Keck Telescope on Mauna Kea is used as a case study. The system is designed to provide low-order wavefront correction using faint natural guide stars. It is shown that two innovations, namely the use of an AO-optimized CCD detector, and use of a variable control bandwidth, result in correction to a Strehl ratio of about 0.1 using guide stars of visual magnitude 18. Sky coverage is approximately 20% at this level of performance. Sky coverage rises to 100% at a Strehl ratio of 0.05.

In this paper we use phase screen models to illustrate the power of the delta rule, by obtaining the optimum reconstructor for a Shack-Hartmann sensor with just 6 subapertures in the form of pie segments. The dependence of the matrix elements and residual error on measurement noise is determined, and the accuracy compared with theoretical limits. Reconstructors for more complex problems involving time dependence and multiple laser spots are ideal applications for the method.

Both laser guide stars and natural guide stars may be used to sample atmospheric turbulence for adaptive optics. The phase representing the turbulence can be expanded as Zernike polynomials across a circular aperture. Assuming a Kolmogorov phase spectrum and C²_n profile the error induced in using off-axis source to correct for atmospheric phase can be calculated for Zernike modes. Small-angle approximations of analytic results are presented and compared with the analytic expressions for model C²_n profiles. A numerical simulation of the analagous errors for the models using an on-axis sodium layer laser guide star is also presented.

The multi-frame blind deconvolution algorithm is considered for processing astronomical speckle images when only a few frames of data are collected. It has been noted that when the speckle data contains even moderate amounts of shot noise the algorithm often converges to the trivial point solution. In this paper we consider a `penalized' blind deconvolution algorithm in which the penalty function is based on the Knox-Thompson algorithm.

We report about the development of novel circuitry for solid state photon counting devices, based on avalanche photodiodes (PC-APD), and tailored for adaptive optics applications. The recent development of EG&G Silk^TM devices, has improved silicon. APDs considerably, reducing the afterpulsing effects, improving the effective Q.E., and reducing the dark current to negligible values. These new APD allow to conceive new quenching circuitry and new applications of solid state photon counters for improved adaptive optics performance.

The requirements of wavefront reconstruction algorithm for use in astronomical adaptive optics systems are discussed. We present the results of realistic numerical simulations of the closed-loop performance of a Shack-Hartmann wavefront sensor used in conjunction with a high-order segmented mirror suitable for adaptive correction on a 4 m telescope at visible wavelengths. Iterative over-relaxation methods are used to derive the segment piston values, and we investigate the behavior of these techniques in the presence of photon and detector noise. We also discuss the use of non variance-minimizing techniques to optimize the point- spread function for particular astronomical applications.

The cost of good, reliable deformable mirrors has made them inaccessible to the astronomical community and other would-be commercial users. Historically, the deformable mirror has cost nearly $DOL2500. per channel and comprised about 25% of the cost of an adaptive optics system. Xinetics has created a deformable mirror technology which costs less than $DOL1000. per channel including the electronic driver. In addition, the mirrors feature novel actuator cartridges which contain a simple thermal mechanism to allow simple replacement of damaged actuators. The current mirrors use piezoelectric actuators for broad temperature response and operate at 100 volts. They feature 7-mm interactuator spacing, 4- micrometers displacement, and interactuator coupling tailorable between 5 to 15%. Complete NASTRAN and Weibull analyses has been conducted to insure sufficient safety margin. The results of the 37, 97, and 349 channel mirror manufacturing efforts are reported in this paper.

LASERDOT has acquired a wide experience in Adaptive Optics Systems: components and hardware manufacturing, dedicated software, system integration and maintenance. Associated with several research organizations, it has taken advantage of having installed the first AO systems routinely used by astronomers. This paper presents deformable mirrors as well as tip- tilt mirror designed and achieved at Laserdot jointly with Observatoire de Paris-Meudon. Some comparisons between the different types of deformable mirrors are done.

We describe the applicability of wave front correction by using a bimorph mirror in conjunction with a curvature sensor. We use Zernike polynomials to describe analytically the quality of the correction for atmospheric turbulence. The match is limited by boundary conditions of the mirror and by the discreteness of the electrodes. The correction is limited by coupling of lower and higher order Zernike polynomials and necessitates an interfacing computer between the wave front sensor and the bimorph mirror.

We report on the design and fabrication of a high bandwidth interferometer suitable for real- time measurement of the figure of a deformable mirror. The design allows for measurement of mirror figure in terms of optical path differences (OPD) between the surface of the mirror and a static reference wavefront. Measurements are made on a 31 by 31 square grid. This instrument is relevant for atmospheric adaptive optics systems because it provides a method for accurately monitoring the figure of a deformable mirror during real-time compensation of atmospheric turbulence. Measured values of OPD on the mirror surface are output in digital form at approximately 10 kHz and can be used as a feedback signal in a digital control-loop for driving the deformable mirror. The system uses a common 4-bucket or 4-measurement interferometric algorithm to compute OPD. The maximum measurable OPD is +/- 7.5 waves. Tests of the completed interferometer indicate that it can routinely measure the dynamic changes in figure of an optical mirror. Preliminary tests indicate that the measurements are accurate to approximately (lambda) /25.

Two different quasi-hexagonal implementations are presented and a variety of mathematical geometries for each are investigated and compared to the WCE/Fried geometry via noise gain analysis.

The bandwidth of a servo controlling a deformable mirror should be determined by the time delays introduced by integration, read out, and analysis of the signals recorded by the wavefront detector. However, many types of piezoelectric actuators used in the construction of deformable mirrors exhibit considerable hysteresis which destabilizes the servo and requires a reduction in the stable system bandwidth. The effect of various degrees of hysteresis on the overall servo performance will be discussed together with methods of its compensation. These methods include `passive correction,' in which the effect of hysteresis is modeled by the servo system and `active correction,' in which the shape of the mirror is continuously measured with a wide bandwidth servo and is used as the error signal. We discuss both types of correction and present experimental results.

The prototype membrane deformable mirror for the Cassegrain adaptive optics of SUBARU (Japanese 8 m telescope under construction at Mauna Kea). It was made of 2 micrometers thick nitrocellulose, 50.8 mm in diameter, and had an effective area of 25 mm in diameter. The surface was coated with aluminum that acts both as a reflective surface and as an electrode. The deformation of mirror surface was produced by applying electrostatic force between the membrane and adjacent multi-element electrodes. The mirror surface quality was 0.03 (lambda) rms at 0.632 micrometers in wave front with and without applying bias voltages. The strong resonance was observed under vacuum condition at 1.6 kHz, but disappeared when the inner pressure was controlled at about 10 Torr, for which the mirror bandwidth became 3.3 kHz at 45 deg phase delay.

Nematic Liquid Crystal Adaptive Phase Screens (NLC-APS) may be used as wavefront correctors for adaptive optics in astronomical uses, provided special techniques are applied to improve their time performance. The advantage of LC-based wavefront correctors are mechanical stability (no moving parts involved), avoidance of mechanical resonances, potential low-cost, and optically transmissive corrections for multiconjugate AO after the telescope focal plane. We do not deal in this paper with other types of LCs which could be used for such purpose, because we find them less favorable, and applicable only to very special observing modes, if any. We have described previously the theoretical and experimental results that have led us to this conclusion. Our goal here is to study the feasibility and the performance of such a wavefront corrector when it is inserted in an adaptive servo-loop, and driven by a 16 X 16 Shack-Hartmann sensor and the reconstructor electronics. The study is carried out via a complete simulation of the system, at the 3.58-m Telescopio Nazionale Galileo (TNG), to be built on La Palma (Canary Islands).

This paper discusses the fabrication considerations and proposed testing concepts for a twelve meter, graphite-epoxy space truss that is being developed to provide structural support for the primary mirror system of the SpacE Laser ENErgy (SELENE) Beam Transmission Optical System (BTOS). A general description of the mirror support configuration is presented. Specific issues which are addressed include low-cost fabrication techniques utilized in the support structure. Later, a description of the dynamic testing program for the entire active primary mirror system is outlined.

A large deformable mirror with 144 actuators was tested in a center-of-curvature configuration. The influence function of an actuator, i.e., the change in wavefront error produced by a unit command to that actuator, was measured interferometrically and compared to predictions made by a finite-element model of the mirror. Good agreement between the measurement and the prediction in both magnitude and shape was found, serving to validate the finite-element model. The measurement process, the finite-element model, error sources, and the dependence of the influence function on the electrostrictive actuator response are discussed.

The design of a proposed 1-m diameter deformable secondary mirror is described, with emphasis on optimization of each component for adaptive optical correction. Finite element analyses are presented which investigate the ability of the mirror to match low-order Zernike polynomials. Wavefront data from a 1-m telescope are used to examine the ability of the mirror to fit atmospherically distorted wavefronts. Finally, mirror surface FEA results from the atmospheric wavefronts are Fourier transformed to predict pre- and post-correction image quality.

The design of two holosteric configurations have been optimized for a maximum center/edge deflexure of 400 micrometers . Their thickness distribution is given for active zones of 16 mm in diameter. The curvature action is obtained from an air-pressure chamber that generates onto the rear side of the mirrors (1) a uniform pressure up to 9 Atm or (2) a central force up to 11 daN. The control of the curvature is made by an accurate pressure gauge. Some preliminary results are shown as obtained on metal prototype VCMs from a first fabrication cycle as well as an X-ray device for testing the machining validity of the boundaries at the edge of mirrors.

In mid 1992 an actively controlled infrared (IR) reflecting secondary mirror was installed on the W.M. Keck Ten Meter Telescope, currently the world's largest telescope. This 0.5 m gold plated beryllium mirror is capable of extremely precise motions over a wide range of amplitude and frequency and is used to perform spatial chopping, a technique commonly used in IR astronomy to improve the signal-to-noise ratio of weak IR sources. This paper describes the design, implementation and test of the mirror drive and control system. Details of the drive electromechanical design and operations on its unique adaptive control system. Test data from laboratory and telescope operation are also shown and discussed, and a short video of the mirror system in operation (shown at the conference) is also briefly described.

The Micro-Precision Control/Structure Interaction (CSI) program at JPL is chartered to develop the structures and control technology needed for sub-micron level stabilization of future optical space systems. The extreme dimensional stability required for such systems derives from the need to maintain the alignment and figure of critical optical elements to a small fraction (typically 1/20th to 1/50th) of the wavelength of detected radiation (about 0.5 micron for visible light, 0.1 micron for ultra-violet light). This (lambda) /50 requirement is common to a broad class of optical systems including filled aperture telescopes (with monolithic or segmented primary mirrors), sparse aperture telescopes, and optical interferometers. The challenge for CSI arises when such systems become large, with spatially distributed optical elements mounted on lightweight, flexible structure. This paper will present an overview of the approach that is being taken by JPL's CSI program to address this challenge. In particular the paper will discuss the application of CSI technology to a specific example of a future large optical space mission. Experimental demonstration of the technology on ground-based testbeds will also be presented.

A correctability model of a large deformable mirror in a center-of-curvature test configuration is developed using finite-element techniques. The mirror is a one-meter diameter, solid ULE facesheet with electrostrictive actuators. The correctability model allows simulation of the performance of the active optical system, as measured by the wavefront error (optical path difference) over the aperture, in response either to commands to the actuators or to noise/disturbance processes. The performance of the closed-loop active optical system, assuming interferometric measurement of the wavefront error, is analyzed as a function of the number of actuators, the breadth of the actuator influence functions, and the spatial content of the disturbance aberration. The measures of performance used are: the correctability ratio, the required actuator stroke and the forces applied to the mirror.

ADONIS is an adaptive optics user friendly instrument for the ESO 3.6-m telescope. The modal control technique is presented as a tool for optimizing the control of ADONIS. The perturbed wavefront is spread over a base of modes treated by the system in terms of their behavior with respect to the observing conditions (turbulence characteristics, anisoplanatic angle and star magnitude). This optimization and the multiple parameters influencing the system performance are difficult to handle by the astronomer. Moreover, the wide range of situations the astronomical AO system has to deal with requires a flexible configuration which has to be optimized for each case. The ADONIS instrument also has to be easy to use by an astronomer without who doesn't have a deep knowledge of the AO techniques. Finally, it must make an efficient use of the telescope time. Hence, a new human interface and an intelligent control system are designed.

The University of Hawaii experimental adaptive optics system is controlled by dual SPARC single board computers on a VME backplane. One processor is dedicated to the feedback loop. The second processor manages loop data flow to a workstation and transfers new control parameters to the loop processor without stopping the loop. This system facilitates cause-effect analysis of the various system parameters.

This paper reports on experiments in modeling and control of the Advanced Structures/Controls Integrated Experiment.

The problem to be solved is to maintain the best efficiency of an Adaptive Optics System (AOS) during an observing run, versus different seeing conditions and star magnitudes. The paper uses the modal control and partial correction concepts as defined in previous works. It has been demonstrated in these works that the Strehl ratio of images corrected by an AOS can be evaluated by a simple criteria. Such a criteria uses only the temporal covariance of the controls applied to the AO mirror. Using these different results a method to optimize the loop gains is suggested. This method can be applied during real time run of the AOS. It uses the only knowledge of the modal control values applied to the AO mirror and the knowledge of the residual wavefront errors as measured by the wavefront sensor.

The Phased Array Mirror, Extendible Large Aperture telescope has been fully assembled and testing has started. The telescope is the first to have a fully adaptive primary mirror, which consists of 36 hexagonal injection-molded Pyrex segments that are seven centimeters flat-to- flat. The segments are mounted on three long-throw voice-coil actuators for tip, tilt, and piston motion. The segment tiles are measured with a Hartmann-Shack wavefront sensor and the piston errors between adjacent segments are measured via inductive edge-sensors. The personnel at NASA MSFC are performing a significant amount of testing in the area of controls/structure interactions; therefore, in addition to a description of the optical performance and aberration correction capability of the telescope, a description of the plan to model the mechanical structure with emphasis on how this will interact with the adaptive optics system is presented.

In this paper we characterize and illustrate Adaptive Optics Control system errors associated with sensor misregistration and present an approach to minimize them by employing nonlinear adaptive control methodologies. Furthermore, through the use of the Extended Kalman Filter methods, we determine if it is feasible that experimental data can be used to drive a complete multi-input multi-output dynamic model of the closed-loop adaptive optics system to estimate parameters directly related to stability margins. Finally, we review some results from an experimental verification of these theories. A closed loop adaptive optics breadboard is used to compare the observed effects of misregistration and other systematic error sources with the predictions of the models described above.

The performance of a closed loop adaptive optics system may in principle be improved by selecting distinct and independently optimized control bandwidths for separate components, or modes, of the wave front distortion profile. In this paper we outline a method for synthesizing and optimizing a multi-bandwidth adaptive optics control system from performance estimates previously derived for single-bandwidth control systems operating over a range of bandwidths. Numerical results are presented for use of an atmospheric turbulence profile consisting of a single translating phase screen with Kolmogorov statistics, a Shack-Hartmann wave front sensor with 8 subapertures across the aperture of the telescope, and a continuous facesheet deformable mirror with actuators conjugate with the corners of the wave front sensor subapertures. The use of multiple control bandwidths significantly relaxes the wave front sensor noise level allowed for the adaptive optics system to operate near the performance limit imposed by fitting error. Nearly all of this reduction is already achieved through the use of a control system utilizing only two distinct bandwidths, one of which is the zero bandwidth.

Planetary detection using doppler shifts requires very high resolution spectrographs which are stable over periods of years. Single or low order mode optical fibers have advantages in this application since they produce spatially scrambled Gaussian input illumination and diffraction limited size input image for spectrograph slit. Very high resolution spectroscopy is thus possible with modest dimensions of the spectrograph. This results in greatly lowering the cost of spectrograph and increasing the stability. However, adaptive optics system is needed to improve the light coupling efficiency. This paper reviews the potential of this technique and presents some preliminary laboratory experiments with the Wavefront Control Experiment (WCE). Plans of future experiment after installing the WCE in a Coude room of a 40 degree(s) reflector telescope at Yerkes Observatory are also presented.

ADONIS (ADaptive OPtics Near Infrared System) is an upgrade of the COME-ON+ adaptive optics prototype. It will allow the astronomical community to use adaptive optics as a common user instrument. This paper describes the main features of the new system, including a mechanical and optical interface for specific visitor equipment and imaging capabilities. We present here the 128 X 128 infrared imaging camera, covering the 1 - 5 micrometers spectral range.

An adaptive optics system for small to medium astronomical telescopes has been designed and is under construction. This adaptive optics system corrects the five lowest order Zernike wavefront aberrations, providing near diffraction-limited performance for telescope diameters smaller than about 5*r_o, and useful correction up to about 10*r_o. This adaptive optics system design consists of a lens-based wavefront aberrator, a CCD-based Shack-Hartmann wavefront sensor, and a personal computer-based feedback loop.

We report on an ongoing experiment at the Starfire Optical Range to study anisoplanatism and investigate a method of mitigating its effect on adaptive optics systems. We have collected Hartmann sensor wave front measurements on binary stars having a variety of separations and have reconstructed wave fronts based on these measurements. At the same time the wave front data was collected, we measured the C²_n profile using a balloon-borne instrument. Using these data, we plan to study anisoplanatism in terms of the structure function and Zernike polynomials. We also plan to investigate whether wave front compensation can be improved by moving the wave front corrector from the aperture plane to a plane determined from the refractive index profile.

Adaptive optical (AO) systems offer large gains in performance for many astronomical applications. We argue here that it is in stellar interferometry that a low-order AO system can be exploited most easily and with greatest effect. A key component in the utilization of partial adaptive correction in interferometers is the use of single-mode fibers to `filter out' the effects of the uncorrected wavefront aberrations. We show that, in addition to making visibility calibration independent of the spatial seeing fluctuations and of the performance of the AO system, fibers provide substantial gains in the signal-to-noise-ratio of fringe visibility measurements when compared with the same adaptive system without fibers. These gains are afforded in both the photon-rich and photon-starved regimes. We calculate the optimum aperture diameters to use with a given order of spatial correction.

We present a simple procedure which estimates aberrations in the exit pupil from point source images taken at the compensated focus. The procedure is based on an iterative technique described by J. Fienup. The measured point spread function of the system produced with a reference source at the input to the adaptive optics system, and the exact, properly scaled shape of the (centrally obscured) exit pupil is all that is needed for the procedure to perform. The results can be obtained quickly, e.g., as part of a daily maintenance procedure. We present some sample cases which were produced with the Come-On Plus adaptive optics system at the ESO 3.6 m telescope on La Silla.

Our group in the Astronomy Department at the University of Illinois is assembling an adaptive optics system that will be installed at the Coude focus of the Mt. Wilson 2.5-meter telescope. The aim of the program is to provide the means to test experimentally, at an astronomically excellent site, many of the newer concepts in laser guided adaptive optics. To accelerate the experimental effort, the system will be built, in part, from existing hardware and from design information provided by U.S. military research groups. Also in the spirit of expediency, the first laser guide star system to be installed at Mt. Wilson will be a single Rayleigh beacon produced by an excimer laser which also exists and has been tested elsewhere. The program as described here is neither final nor fixed because the hardware, the technology, the system configuration, and the ideas that drive it will continue to evolve during the five year duration of the project.

The COME-ON+ adaptive optics system was set up on the ESO 3.6-meter telescope for two technical runs (in December 1992 and April 1993) and is now routinely used for astronomical observations. This system is an upgrade of the COME-ON adaptive optics prototype system. During the technical runs, images were recorded in the V, I, J, K and L' spectral bands. Currently, the best resolution obtained is 0.1' in H (1.65 micrometers ) under bad seeing conditions (seeing > 1' and averaged wind speed > 10 m/s) and reference star magnitudes of 6 to 8. The corresponding Strehl ratio is 35%. 70% Strehl ratio was obtained at 2.2 micrometers (K band). At 0.9 micrometers (I band), in the partial correction regime, the resolution is of the order of 0.2' for 0.8' seeing and 10 m/s averaged wind speed. The optimized modal control has been used on faint reference stars. The limiting magnitude (in V band) for wavefront sensing has been measured to 14 and 15 depending on the spectral type of the reference star and the seeing conditions for a low frequency tip-tilt correction.

Since 1990, an adaptive optics system with 52 actuators, a sensitive wavefront sensor and two near-infrared cameras equips the 3.6 meter telescope at La Silla (Chile). Open to visitors, this instrument has already produced a number of demonstrative and important scientific results in planetary, stellar and extragalactic astronomy, where the diffraction limit of the telescope is reached and routinely leads to images with 0.1 - 0.2 arcsecond resolution in the near infrared ((lambda) > 1 micrometers ). This instrument is a prototype developed in view of the European Very Large Telescope. We present here an overview of the science already done with some prospects for the near future.

A prototype adaptive optics system has been developed at Lawrence Livermore National Laboratory for use at Lick Observatory. This system is based on an ITEK 69-actuator continuous-surface deformable mirror, a Kodak fast-framing intensified CCD camera, and a Mercury VME board containing four Intel i860 processors. The system has been tested using natural reference stars on the 40-inch Nickel telescope at Lick Observatory yielding up to a factor of 10 increase in image peak intensity and a factor of 6 reduction in image full width at half maximum. These results are consistent with theoretical expectations.

The Wavefront Control Experiment is a high temporal bandwidth visible wavelength adaptive optics system on loan at the University of Chicago by the Ballistic Missile Defense Organization for astronomical research. In the last year numerous laboratory tests have been conducted with the WCE. The system will soon be installed in a coude laboratory associated with a Ritchey-Cretien 40' telescope at Yerkes Observatory. A brief overview is given of the WCE system, our recent laboratory tests with different matrix reconstructors, the 40' telescope and coude beam optics, and future plans for the instrument. Our scientific plans for the WCE include monitoring the atmosphere of Jupiter prior to and after the projected impact by comet Shoemaker-Levy in July 1994, and to implement a long term program to do high resolution single-mode fiber feed spectroscopy for asteroseismology and extra-solar planetary detection, via radial velocity determinations, of bright stars.

This paper describes the method used to determine the actuator strokes applied to a deformable mirror to achieve planet detection and the design of a coronograph which implements this novel technique.

In recent years there were published some papers dealing with an important problem on the use of adaptive means and adaptive optical systems to improve the images and beams observed through the atmosphere.

The point spread function (PSF) for a ground-based telescope is calculated for two types of wave front correctors, i.e., deformable and segmented mirrors. The deformable mirror is assumed to be a thin square plate fixed at its center and deformed within a system of lateral forces applied to 20 points. The segmented corrector is a set of square segments with three degrees of freedom, the number of elements varies from 1 to 16. The wavefront distortions are of the Kolmogorov spectrum and are considered to be known for each random realization. It is shown that the PSF calculated for a 9 element segmented corrector (27 degrees of freedom) is close to that obtained for the deformable mirror with 20 degrees of freedom.

The basic principles of nonreciprocal optical systems with phase -conjugating mirrors design and their main properties are presented. Such system can form high quality images by means of telescope with distorting large primary mirror. The demands to quality of the mirror are hundreds of times less strict, than for the common type telescope. Several different realizations of such systems were investigated experimentally and numerically.

The results cI experimental realizaticu of the 1as telescce with the phase-axijugation axnpensaticii 1 segmented main mirrcr w ith the diameter 300 mm, focal length 1200 mm are presented. The expiment results, which are correlated witti the numerical model results, show the pcssibility of diffraction limited image formation while segments angle irusmatch is up to 13X 1O and step mistake up to several dozen of micrometers.

Control system of segmented mirror is being researched for case of spatial element location independent testing relative to common base. There is developed the mirror segment model as control subject in analytic form, which takes in account the mirror geometric features. Statistical accuracy problem for multi-dimensional regulator is solved by means of some relations obtained which allow to optimize actuators location using criteria of interchannel crosstalk elimination in control system and minimization of spatial position errors of the mirror segments.

A laboratory prototype of the nematic liquid crystal (LC) phase corrector for adaptive telescope was developed. The corrector consists of a LC transparent controlled by light signal and a controller, transforming input data of the telescope wavefront distortion into proper light signals.

Experimental investigations and numerical simulation of the thermal control system performances as well as local electric heaters control parameters are discussed. The possibility of thermal stabilization of light-weight segment for fused silica mirror is investigated. Heating and cooling times for the structural mirror are calculated. Temperature fluctuations are modelled for the mirror surface by means of a lamp heat radiator. Capabilities of the Thermo- Optical Compensation System analyzed.

An effort is made to synthesize speckle interferometric (SI) imaging technique, which is adapted (immunized) to arbitrary type of isoplanatic multiplicative distortions and does not count on usage of a reference sourse (RS), auxiliary wavefront sensor (WFS) and a priori information concerning the object being observed: all necessary imagery information is drawn from distorted image snapshots (specklegrams), total number of which is arbitrary (N 2). The proposed method makes possible the fusion of data acquired with various spatially separated equally-scaled optical systems affected by essentially different perturbing phase distortions within unified scheme. The procedure is based on Projection onto Convex Sets (POCS) approach and, in contrast to Blind Deconvolution (BD), enables the iterative estimate of OTF to be found only through the set of snapshots available and without appealing to current estimate of unknown object brightness. The solution uniqueness is briefly analysed. Generalizations of such SI-methods as Labeyri's Processing (LP), Knox-Thompson (KT) SI and Triple Correlation (TC) method are derived. The way toward possible algorithmic combining the data acquired with both an object under observation and arbitrarily corrupted RS (even though it may have unknown resolvable spatial brightness) is discussed. This aspect of proposed technique is important in the light of temporary difficulties associated with the high-quality Laser Guide Star (LGS) generation. Field results on unreferenced restorating the images of binary star and Orbital Complex (OC) consisting of Space Station "Mir" and "Soyuz TM" spacecraft are presented. Pursuance of research is motivated by desire for increasing the autonomy and universality of telescopes at their reasonable cost, what is essential for solving both astronomical and radar-aided space control problems.

Techniques and instruiients for testing and both preliminary and fine phasing of segiiented aiirror and its segiients are proposed. These techniques and instruments are characterized by high accuracy, compa-' rative simplicity and low cost.

An uncooled, 293 actuator deformable mirror design has been developed and two mirrors have been produced. This design is designated the High Performance Active Corrective Element (HiPACE) 293. The HiPACE 293 will provide substantial wavefront correction for large, multi-meter telescopes operating at science wavelengths of >= 2 micrometers while providing correction at shorter wavelengths for facilities with more moderate apertures.

Optical interferometric high stability fields produced by laser beam are proposed to be used as an optical coordinate system. To produce such interferometric field over the large surface a set of optical elements and systems are proposed which consist of surfaces formed by means of off-axis rotation of parabola, ellipse or hyperbola and their confocal combinations (one of two focuses of each curve is common). If the source is situated at one of focuses (but not at the common one) such systems are completely free of backward reflected beam and aberrations. These systems allow one to reach the extremely small beam divergence and, hence, to get the extremely small focal spot. It is very important in some special applications. Such elements can be also independently used in laser optics, adaptive optics, interferometry, for optical and radio telescopes and antennas and so on. The holographic methods may be helpful in technology of manufacturing the above elements.

This paper investigates and compares wavefront sensing options for a proposed natural guide star adaptive optics upgrade for the Keck telescope. The methods of curvature sensing and slope sensing with a Hartmann sensor are compared. A curvature sensor model is developed that includes nonlinear, diffraction, and noise effects. The analytical model is validated in simulation using the Keck optical prescription parameters. A comprehensive treatment of estimation/reconstruction is presented and used for the Hartmann sensor/curvature sensor comparisons.

The Canada-France-Hawaii Telescope is developing an Adaptive Optics Instrument Adaptor for general use at its 3.6 m telescope. It consists of a 19 electrode, bimorph deformable mirror (1:6:12) used with a 19 sub-aperture, curvature wavefront sensor.

We report on the results of the system simulation software written at Canada-France-Hawaii Telescope, to predict and optimize the performance of our adaptive optics bonnette presently under construction. The individual simulation elements, atmospheric simulator, curvature sensor, bimorph mirror and control loop, are reviewed with an emphasis on the basic properties of curvature sensors. Optimization of the extra-focal distance parameter and its consequences are discussed. We then present results of whole system simulations, and quantify the main sources of error. Some results, including noise, are reported. In a second part, we present our view of modal control and detail the construction of the modal basis. We also report briefly on modal gain optimization.

Conventional modal compensation using Zernike polynomial expansion by means of Hartmann-Shack wavefront sensors is discussed. Cross-talk problems of Zernike coefficients have been found to be serious, especially for large sub-aperture configurations. A proposal for use of a modified Hartmann-Shack wavefront sensor and iterative wavefront reconstruction is presented and a comparison is made between conventional and proposed methods.

A method is described for production of time series of atmospherically distorted wavefronts following Kolmogorov statistics. The method uses Taylor's hypothesis (`frozen turbulence') to generate the temporal development of the aberrations caused by an atmospheric turbulence layer. An arbitrary number of turbulence layers can be used to generate the final wavefront.

In order to increase the corrected field of view in multi-conjugate adaptive optics we suggest a method for measuring and separating the contribution of atmospheric turbulent layers. The sodium layer serves as a huge screen, on which we project a wide fringe pattern from a single laser on the ground. A modified Hartmann-Shack sensor is employed to detect deformations in the pattern: sections of the fringe pattern are imaged by a lenslet array onto a large-format camera. Low layer turbulence causes overall shift of the fringe pattern in each lenslet, while high altitude turbulence results in internal deformations in the pattern, which are detected by that section of the camera which is behind each lenslet. Parallel Fourier analyses for the different lenslets allows separation of the atmospheric layers: the periodicity of the fringes lends itself to digital demodulation, which yields the deformations in the fringe. We present a statistical error analysis and simulations, showing good performance over a field of view of 80', compared to the performance expected of a conventional single sodium beacon, single adaptive mirror system.

We investigate the feasibility of using an adaptive mirror for in-orbit aberration corrections. The advantage of an in-situ aberration correction of optical components in the space environment is that the mirror shape can be adjusted in an iterative fashion until the best image is obtained. Using the actuator spacing, corresponding to one-half the Nyquist frequency, the Strehl ratio of the corrected wavefront improves to 0.95 when the mirror is fabricated with 6.5 waves of spherical aberration. The Strehl ratio is decreased to 0.86 when the number of actuators is reduced by a factor of 4, in a 2D adaptive optics model.

Wavefront sensors have been used to make measurements in fluid-dynamics and for closed loop control of adaptive optics. In most common Shack-Hartmann wavefront sensors, the light is broken up into series of rectangular or hexagonal apertures that divide the light into a series of focal spots. The position of these focal spots is used to determine the wavefront slopes over each subaperture. Using binary optics technology, we have developed a hierarchical or fractal wavefront sensor that divides the subapertures up on a more optimal fashion. We have demonstrated this concept for up to four tiers and developed the wavefront reconstruction methods for both segmented adaptive optics and continuous wavefront measurement.

A theory for a new laser guide star technique is developed. This technique, for the first time, permits the sensing of a full-aperture tilt of the atmospheric wave front distortions using a laser guide star, eliminating one of the fundamental limitations of adaptive optics. By using a detailed analysis of the laser guide star image jitter for a conventional scheme and for its possible modifications, it is shown that, because of reciprocity of propagation paths, a conventional laser beacon is unable to sense a full aperture tilt and can be only used to measure the higher-order wavefront distortions. The full aperture tilt can be measured with a divergent beam if its effective size at a laser beacon altitude coincides with the radius of a receiving telescope. Estimates of the modified laser guide star image jitter rms, signal-to-noise ratio, and the mean square deviation between the instantaneous tilts for the modified laser beacon and a natural star are obtained, and it is shown that the proposed technique is practical.

A variety of clever pre- and post-detection methods have been developed to image through atmospheric turbulence. These methods have been developed and exercised primarily for use in the isoplanatic (space-invariant) imaging case. Multi-conjugate methods have been investigated for the accommodation of space-variance with compensated imaging, however multiple guidestars, wavefront sensors, and deformable mirrors are required making a practical implementation uncertain. Post-detection correction methods can be extended to the space-variant case with mosaicing methods, but such approaches provide suboptimal estimates.

Results of wavefront control and a simple example of wavefront correction using a 64 X 64 pixel nematic liquid crystal array are presented.

Experimental results or new approach to the problem oI sharpness runct ion maximization in adaptive optics are presented. The main dirre.. rence rrom ordinary method is in rorming of signals, proportional to the components or sharpness runction gradient, in optical proces sor. This solution is very errective in case or low light intensity, when the quantum errects of registration are essential. It is shown that in particular case the scheme or adaptive system is similar to curvature sensing system. Experimen tal system provides 32—zonal correction or wave front in real atmospheric conditions for objects of various forms. Subject terms: adczpttve opttcs; deformabLe mtr rors; shc1rpness funct ton; phase dtstort tons.