A performance prediction model for ground-based adaptive optics systems has been developed which permits the optimization of the spatial and temporal system response at a fixed throughput. The temporal response of the system is approximated to be that of a single-pole filter with a time constant (tau) . Approximate fits to actuator influence functions are used to model the adaptive mirror response, which is included as a high-pass spatial filter of the incident wavefront. The incident wavefront is also assumed to be spatially filtered by the sampling apertures of the wavefront sensor. The computed rms deviation of the corrected wavefront is used as the system performance measure. This system performance is parameterized in terms of the throughput a²v(tau) , where a is the actuator spacing and subaperture size and v is the pseudo-wind speed characteristic of the turbulence producing the wavefront aberrations, and the ratio of the temporal and spatial sampling intervals v(tau) /a. For large aperture telescopes with D >> r₀, the performance curves are found to be independent of D and the optimal spatial and temporal bandwidths may be readily selected for the desired throughput, this parameterization provides a simple means of evaluating total system performance from typical wavefront sensor noise models.

One method for scaling lasers to higher power is to build several separate amplifier chains, and then coherently combine the individual beams together. To combine the beams the pathlengths must be matched to an integer number of waves, and the tip and tilt of each beam must be the same. Since optical tolerances are fractional micrometers, a sophisticated control system must be employed to actively measure tip/tilt and piston errors, and apply real time corrections. We have developed a test bed that allows us to develop the required control algorithms for two to four beams and then scale to a larger number.

We are developing an adaptive optics system that is based on an array of actuators arranged with subapertures that are equilateral triangles. The wavefront sensor is a video Hartmann sensor that also uses an equilateral array of lenslets. The controller hardware uses a VME bus. The design minimizes the generation of reflected wavefronts higher than first order across each lenslet for large excursions of actuators from positions where the mirror is flat and, thus maximizes the precision of the slopes measured by the Hartmann sensor. The design is also immune to the waffle mode that is present in the reconstructors of adaptive optics systems where actuators are arranged in a square array.

The design of an Adaptive Optics System must take into account at least the following main criterias. First, correction efficiencies versus the atmospherical turbulences, i.e., the Fried diameter and the turbulence power spectrum. Second, sky coverage versus star magnitudes and star classes. To answer these criteria an optimization method is defined. It uses as main parameter the residual mean phase errors of the corrected wavefronts. It is demonstrated that this parameter able us to optimize the System performances: number of modes to be controlled, sampling rates of phase measurements, and noise minimization. Finally an application is done for a 3.6 m class telescope and 2 different adaptive mirror types.

We report on the operation and performance of a complete integrated 1 m adaptive optics systems for compensation of atmospheric distortion of optical wavefronts. Both visible artificial laser guide stars (doubled Nd:YAG laser with wavelength of 0.532 micrometers ) and natural stars can be used as sources for reference wavefronts. A polarization shearing interferometer which uses a narrow optical bandwidth and has 500 subapertures is employed to sense wavefront distortion. These measurements are used to compute a conjugate wavefront to the distorted input light. The computed conjugate is then imprinted on a deformable mirror which consists of 500 individual square mirror segments. The effectiveness of the compensation is determined from a measured PSF of the system. Both indoor benchtop and atmospheric experiments are under way to test the performance of the integrated system. The results of these tests so far are very promising, yielding short-exposure images at 0.532 microns which contain discernible energy at the diffraction limit of 0.1 arcsec.

Recent results from the Laser Guide Star Project at Lawrence Livermore National Laboratory are presented. Photometry of the return signal has shown that the photon return is approximately 10 photons/cm²/ms at the pupil of the receiving telescope in agreement with a detailed model of the sodium interaction. Wavefronts of the laser guide star have also been measured with a Shack-Hartmann technique and power spectra have been shown to agree with those of nearby natural stars. Plans for closed loop demonstrations using the laser guide star at LLNL and nearby Lick Observatory are discussed.

We have developed an adaptive optics image compensation system. It consists of a tip-tilt mirror, a 21 actuator deformable mirror, a photon counting shearing interferometer, a digital wavefront processor and an ICCD array camera. In this system, both the wavefront sensing and the correction are performed at visible wavelengths. The servo bandwidth of the system in open-loop is 65 Hz at 0 dB point. This system has been successfully tested on a telescope of Yunnan Observatory in China. The tested aperture is 375 nm. The system works satisfactory with stars brighter than m_v equals 3^m.78. The minimum photocount of wavefront sensing is 95 per subaperture per integrated time. The experiment results is presented, which demonstrate the considerable gain in resolution either for single stars or for binary stars.

The electro-mechanical equations governing the motion of the surface of a deformable mirror, due to the activation of a piezoelectric or electrostrictive actuator mechanically located between the mirror surface and the mirror substrate, are derived in terms of the fundamental electro-mechanical parameters of the mirror-actuator configuration. The equations of motion are solved in the frequency domain and the solutions show reasonable agreement with experimental data such as Bode plots, which characterize the frequency dependence of the amplitude and phase of the mirror surface motion, and with frequency dependent impedance measurements. Time domain solutions are used to construct time histories of an actuator under a step-voltage input.

Laserdot is a company which manufactures both Stacked Array Mirrors (SAM) and Bimorph mirrors (BIM) types. The first type is well known as the COME-ON plus type, working on European Southern Observatory (ESO) telescope, and the second one is working successfully on University of Hawaii bench, lead by Francois Roddier. As it does not yet exist exhaustive comparison between both mirrors LASERDOT finds interesting to compare them in full objectivity to know which kind of mirror is better dedicated to one or to the other application. Based on our recent works, the two components are compared and we give a few criterions which, depending on the application field and goals, make easier the choice between both technologies.

The low voltage SELECT deformable mirror described herein provides sub-nanometer control of the optical wavefront and represents the culmination of 10 years of development. Due to the maturity of the PMN electrostrictive actuator technology, the mirror operating voltage has decreased from 3000 volts to 100 volts, the hysteresis has been reduced from near 20% to less than 1%, and the response uniformity is better than 3% over the entire actuator population. But more importantly, the deformable mirrors built today are unparalleled in terms of performance and reliability, even after millions of operational cycles. This paper will summarizes the actuator process developments and discuss the design features and measured performance of the low voltage SELECT deformable mirror family. In fact as the results will show, the SELECT DM's are not only good deformable mirrors, but are also among the most precise optics ever made, especially when operated under active control.

A large aperture, liquid cooled Deformable Mirror (DM) is being developed in support of the ALPHA-LAMP Integration (ALI) experiment requirements. This development will result in a continuous facesheet DM having an active aperture of greater than 28 cm, which is believed to be the largest diameter operational unit developed to date. The mirror is actuated in the active and surrounding guard ring regions by 241 low voltage piezoelectric actuators with a full stroke capability of 10 micrometers for interactuator strokes of up to 2.5 micrometers . The coupling between an individual actuator and the nearest neighbor actuators is projected to be approximately 12%. Performance tests on a smaller scale, simulator unit indicate that all requirements for a high optical power DM subsystem will be met. The design, fabrication, and simulator signal testing phases have been completed and the intermediate results will be reported. The mirror and drive electronics integration and testing are scheduled for completion later in 1993. Performance characteristics of the simulator and portions of the final mirror and drive electronics will be presented, including transfer function, gain and phase data, measured influence function, phase map comparisons with modeling predictions, a 2-dimensional analytical form for the influence function, and surface figure characteristics of the completed mirror.

We describe the motivations for building low-cost deformable mirrors constructed with the needs of astronomers in mind. We detail the design and manufacture of a 59 actuator continuous faceplate deformable mirror and its associated driving electronics and give a summary of the initial test results of this mirror.

Adaptive optics systems have proved their efficiency to obtain high resolution images in he field of large astronomical telescopes. The same technics can be applied to correct X-ray mirror shapes. The paper describes the principle of an adaptive X-ray mirror system (mirror architecture, measurement subassembly, control unit). In a second part, first results obtained during the design study of the ESRF adaptive X-ray mirror are given. The possibility to achieve cylindrical or elliptical mirror surfaces using adaptive optics technics are suggested.

A large-aperture, reactionless Fast Steering Mirror (FSM) is described and modeled, and performance data are presented. The FSM optical element is 24 cm in diameter and is capable of +/- 3 mrad angular stroke in both azimuth and elevation. Dedicated position sensors provide feedback to enable two-axis closed-loop operation at frequencies greater than 1 kHz. The system is modeled as a set of coupled differential equations, from which performance predictions can be made. The performance of the system is presented in terms of spectral power consumption, spectral reaction attenuation, and closed-loop behavior. Closed-loop performance is characterized in terms of open-loop, closed-loop, and error attenuation transfer functions. Comparison to predictions from the dynamic model are made where applicable.

Designs for large lightweight spaceborne optical systems are evolving to configurations that utilize segmented apertures. Optical quality and stability requirements for these systems are in general more stringent than today's best telescopes. These two factors require the use of high precision active control systems to correct and maintain optical performance. The Eastman Kodak Company is in the process of developing the technologies required to demonstrate that such systems are feasible. The chosen approach utilizes a structural control system to minimize high bandwidth disturbances and a low bandwidth optical control system to correct and maintain the optical quality of the telescope. An overview of this approach and a description of current progress is discussed.

This presentation describes fast two-axis beam-steering mechanisms known as TABS built at General Scanning over the years. These TABS approximate a two dimensional fulcrum and can be classified by the number of moving optical elements; one, two, or three. Only large motion devices are addressed here. One-mirror small motion TABS have been hotly pursued for SDI applications, and the SPIE Proceedings, Volume 1543, examines a number of them. Conventional gimbals systems have been generously described elsewhere and also are not treated here. In the common gimbals system, the torque motor for the second axis is transported on the structure mounted on the first axis. There are no universally preferred solutions to two-axis beam steering. Each application has evolved its preferred solution. From the point of view of the optics design, large motion devices such as TABS introduce a number of constrains to speed and excursion.

A high performance reactionless scan mirror mechanism was developed for space applications. This paper presents a number of key new developments in the area of mechanical design of large scanner mechanisms. The design incorporates a unique mechanical means of providing reactionless operation which also minimizes weight, mechanical resonance operation to minimize power, combined use of a single optical encoder to sense coarse and fine angular position, and a new kinematic mount of the mirror. Along with the mechanical description, current status of the project is given.

The hardware building blocks of all contemporary wavefront sensors for astronomical adaptive optics can be described as 'some front end optics, one or more CCD cameras, and a processing computer'. Selection of one of these sensors for a new installation should be based on a comparative cost/performance/risk evaluation. There are four levels of evaluation. First, we can calculate a 'phase error per photon' figure of merit inherent to the optical transformation. Second, we can evaluate the effect of various effects on the noise (precision) and accuracy of the sensors. Third, we can examine the complexity of the optical transformation from non-detectable wavefront to detectable intensity pattern and the concomitant processing complexity to extract the phase from the detected intensity. Finally, we can estimate the engineering difficulties in implementing the desired optical transformation. We suggest that the first level of examination, while providing an important, quantitative performance discriminator, does not provide a basis for sensor selection. The second level of evaluation, often approached qualitatively, suggests possible operational limits for the sensors. The third level suggests hidden difficulties, while the fourth level is perhaps a cost or risk discriminator.

A novel adaptive optical system is being developed at NASA's Marshall Space Flight Center (MSFC). The system is based on the idea of integrating the wave front sensor into the mirror of a segmented system. This approach provides a self contained, adaptive optic which can accurately measure and correct for the localized tilt of the incoming wave front. The mirror components contain the sensing elements, the actuators and potentially some of the reconstruction processing. These segments have a hexagonal shape and will be used to form a segmented, adaptive optic, mirror. The potential mission for this type of mirror is in the form of the primary of a 12 meter class beam expander to direct laser energy through the atmosphere. This energy can then be utilized to power orbital transfer vehicles, charge depleted batteries on satellites during solar eclipse or other future space missions.

A 256-channel Digital Wavefront Reconstructor System (DRS) has been developed for the Sacramento Peak Adaptive Optics System. The system may be configured with 16 to 256 channels in a single VME chassis. Several thousand channels can be accommodated by adding more VME chassis. The system architecture supports both parallel and serial processing combinations. This flexibility allows it to be adapted to various wavefront sensors or used in other applications that require high-speed parallel data processing in real time. Algorithms may be implemented that perform matrix multiplication, FIR and IIR filters, gain and offset corrections, table lookups, and polynomial evaluations. The use of the DRS in the Sac Peak Adaptive Optics System is presented, followed by a detailed description of the DRS hardware design.

Hartmann-shack method is widely used in adaptive optics system for wavefront sensing. A data processing system is used in the sensor for computing wavefront gradient, wavefront reconstruction and control. To get a wide control bandwidth, the system speed should be high enough and the time delay should be as short as possible. In this paper, a multiprocessor system for a 37-element Hartmann-Shack Wavefront Sensor (H-S WFS) is presented. The main computation is completed by 4 processing modules in parallel, and the data acquisition, processing and communication are conducted in pipeline. The system can process video signal at 380 frames/sec and the time delay is less than 1/8 of frame period.

Accessing a large field-of-regard (FOR) from an aircraft-mounted infrared system imposes significant structural and aerodynamic penalties. A novel conformal infrared (IR) transceiver concept is presented which is currently under development. A trial design of this concept can access a 160 deg FOR without a gimbal mirror or 'fish eye' lens. A fiber optic bundle is used to allow a wide range of beamsteering technologies with small steering angles (i.e., +/- 5 degree(s)) to access the large FOR (+/- 80 deg) through a single, conformal aperture. The output lens size is less than a factor of three times larger than the input/output IR beam, yet provides near diffraction limited polychromatic collimation over the full FOR. The concept is applicable over a wide spectral band (ultraviolet to far IR), however, it is being developed for the mid-IR (2 - 6 micron) band. The challenging technical aspects of the fiber optics in this spectral band are discussed.

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 aspects of the adaptive reconfiguration capabilities of the Hartmann wavefront sensor to be used for the verification of the Hubble Space Telescope optics. That sensor has the ability to adjust itself to misregistration with the system pupil via the modification of algorithm parameters. It also corrects for misalignment with the image location via active repositioning of the optical head.

An overview of the Wavefront Control Experiment (WCE) adaptive optics system is presented. We discuss our initial tests of the system and illustrate adaptive correction capability of the system using its internal star source simulator. Future laboratory tests are planned prior to adapting the system to a coude feed of a 41' reflecting telescope at Yerkes Observatory where atmospheric compensation with astronomical sources will be investigated.

The next few years will witness the construction of several new 6.5 - 10 meter class astronomical telescopes. Imaging with these large instruments not only requires us to gain better understanding of the nature of atmospheric phase distortions over large spatial scales but should also motivate the adaptive optics community to rethink the ways in which we sense wavefront structure. At infrared wavelengths, direct interference of spatially coherent infrared starlight affords engineers of adaptive optical systems for filled apertures a ruler against which to measure all wavefront sensor designs. Direct interferometry tells use exactly what we want to known, the relative value of the wavefront phase at various points across the telescope pupil. Direct interferometry, implemented by splitting off a small fraction of the available photons using a suitable pupil mask, pins down phase difference values that can be used as verification for any wavefront sensing scheme. One technique investigated toward the goal of the application of interferometric techniques to filled aperture telescopes is described below. In this Fourier transform-based technique, direct interferometry provides phase information for the control of adaptive optics systems without the necessity for an independent wavefront sensor. Here we describe and analyze the noise properties of one technique for faint stellar sources.

The Wide Field and Planetary Camera (WFPC) is the principal instrument of the Hubble Space Telescope (HST), occupying the central portion of the telescope's focal plane. The Wide Field Camera meets the originally conceived requirement for an imaging device that covers a square field of view 2.67 arc minutes on a side with a pixel size of 0.1 arc second. The so-called Planetary Camera of WFPC offers a longer effective focal length over a smaller field (yielding 0.043 arc second per pixel) to better sample the point spread function of the telescope for critical definition imaging. The first generation WFPC (WFPC-1) was initiated in late 1977 and launched with the HST in April, 1990. A second generation backup instrument (WFPC-2) currently scheduled for launch in late 1993 will carry corrective optics to restore the flawed vision of the HST. The present paper traces the history of these developments.

The correctability of the primary mirror spherical error in the Wide Field/Planetary Camera (WF/PC) is sensitive to the precise alignment of the incoming aberrated beam onto the corrective elements. Articulating fold mirrors that provide +/- 1 milliradian of tilt in 2 axes are required to allow for alignment corrections in orbit as part of the fix for the Hubble space telescope. An engineering study was made by Itek Optical Systems and the Jet Propulsion Laboratory (JPL) to investigate replacement of fixed fold mirrors within the existing WF/PC optical bench with articulating mirrors. The study contract developed the base line requirements, established the suitability of lead magnesium niobate (PMN) actuators and evaluated several tilt mechanism concepts. Two engineering model articulating mirrors were produced to demonstrate the function of the tilt mechanism to provide +/- 1 milliradian of tilt, packaging within the space constraints and manufacturing techniques including the machining of the invar tilt mechanism and lightweight glass mirrors. The success of the engineering models led to the follow on design and fabrication of 3 flight mirrors that have been incorporated into the WF/PC to be placed into the Hubble Space Telescope as part of the servicing mission scheduled for late 1993.

A very compact tip/tilt mirror has been developed for the Wide-Field/Planetary Camera II, a science instrument that is to be installed in the Hubble Space Telescope to restore the Hubble's imaging performance. The Articulating Fold Mirror (AFM) is a space qualified, ultraviolet compatible device that incorporates many advanced features including a highly lightweighted mirror and electrostrictive solid state actuators that provide precise and repeatable open loop performance. The design, fabrication, and testing of the AFM are described.

The Articulating Fold Mirror (AFM) for the Wide Field/Planetary Camera-II (WF/PC-II) instrument is a very compact, complicated, highly precise mechanism. The AFM's basic function is to provide tip and tilt correction in the optical paths of the WF/PC-II instrument. Its necessity is brought about indirectly by the spherical aberration of the primary mirror in the Hubble Space Telescope (HST). Many challenges are created by the necessity of the new mechanism in the optical design. (1) The new mechanism must exhibit high precision in the placement of the mirror surface in two rotations (tip and tilt). (2) The available packaging volume for the AFM is very shallow and requires an innovative approach to achieve the necessary performance requirements. (3) The schedule for delivery of the flight certified AFM's is extremely tight, and as such does not allow for any failures during the qualification phase of the AFM project. Structural design and analysis plays a major role in meeting the stringent performance requirements within the schedule and fiscal constraints. The final result is a qualified mechanism which meets or surpasses all of its requirements.

As part of the HST repair mission it is necessary to verify the performance of the correction optics before their installation in the telescope. To accomplish this precision testing a Hartmann style wavefront sensor and pupil parameter measurement tool has been designed and built. This instrument, termed the Aberrated Beam Analyzer (ABA), will be used to measure the wavefront of both aberrated HST simulators and the unaberrated output of the correction optics. In addition, the ABA measures the location, size, and obscuration ratio of the exit pupil of the system under test. Parameters such as the chief ray angle, PSF, MTF, encircled energy, and Strehl ratio are calculated from the measured data. Operation of the ABA is fully automated and is controlled via a high level scripting language. All data is permanently archived on optical disks for later analysis. The design and theory of operation of the ABA will be discussed. Particular emphasis will be given to the error budget and the measurement performance of the ABA. Some preliminary data will be presented.

Low spatial frequencies of atmospheric turbulence are specially troublesome to astronomers because the phase distortions they cause have large amplitude. We have begun experiments at the Multiple Mirror Telescope (MMT) to remove these errors with tip, tilt, and piston control of pieces of the wave front defined by the telescope's six 1.8 m primary mirrors. We show long exposure images taken at the telescope with resolution as high as 0.08 arcsec under piston control, and 0.26 arcsec under tilt control, using an adaptive instrument designed to restore diffraction-limited imaging in the near infrared. We also present preliminary results from analysis of images of the pre-main sequence star T Tauri taken with tilt control of the six beams only, at three infrared wavelengths. The resolution is between 0.35 and 0.4 arcsec, higher than has previously been achieved with direct imaging. The faint red companion to T Tau is clearly revealed, and is seen to be undergoing an energetic outburst.

The Lockheed Missiles and Space Co. (LMSC) has developed a 57-actuator segmented adaptive optics (AO) system to compensate for atmospheric turbulence encountered in ground- based solar astronomy. While working with the National Solar Observatory (NSO), this system has been successfully tested in several observing runs on the Tower Telescope at Sacramento Peak. This paper gives a brief description of the AO system with emphasis on recent developments. Images from a recent observing run with the AO system are also presented.

We have recently developed the first adaptive system to utilize a tip-tilt Cassegrain secondary mirror. The system (called FASTTRAC) is also unique since it uses an infrared array to track the brightest speckle of the guide star. FASTTRAC can correct wavefront slope fluctuations at up to 100 Hz closed loop and stabilize image motion to less than 0.1' rms in the focal plane. The system requires an infrared guide star of 8th mag at K (2.2 micrometers ). The resulting 54 X 54' images show improvements of up to approximately 300% in resolution (FWHM) when tracking the brightest speckle of the infrared guide star.

The Canada-France-Hawaii Telescope has undertaken the development of an Adaptive Optic bonette for general use at its telescope. There will be separate mirrors for tip-tilt and for higher order corrections. A 19 element curvature wavefront sensor will be used with a bimorph (19 electrodes) deformable mirror. Modal control will be used in order to optimize the number of modes corrected according to the atmospheric turbulence coherence time and brightness of the reference source. The optical design minimizes the number of reflections by using off-axis mirrors. It is planned that various instruments such as CCD cameras, IR cameras and integral field Spectrographs (TIGER type) will be used. At this point phase A design studies have been completed and fabrication has started.

We have mounted an early adaptive optics system, the Atmospheric Compensation Experiment (ACE), on the 60-inch telescope at Mount Wilson Observatory in California in a program designed to investigate the performance of ACE at an astronomical site and to evaluate the usefulness of adaptive optics for astronomy. Despite its development as a non-astronomical instrument, ACE has produced positive results, including the obtaining of images of single and double stars with a resolution (full-width half-maximum) of 117 milliarcseconds at 700 nm. Improvement of image quality is obtained for guide objects with a B magnitude brighter than 5.9. To deepen this limiting magnitude, we have embarked on a low-noise high-speed CCD fabrication project jointly with JPL. First devices have been fabricated. We have applied post- processing techniques borrowed from speckle methodology to the adaptive optics images, and find that the pre- and post-processing techniques complement each other powerfully. We conclude that an adaptive optics system designed specifically for visible-wavelength astronomy would be a low-order system with good site thermal control, combined with post-processing. Such a system could be effective, robust and relatively low cost.

Design of different ground-based adaptive telescope is considered. The analysis is based on computer simulation. Here I consider principles of computer simulation of the wave-front, distorted by atmospheric turbulence, including problems of large scale simulation and dynamic simulation of wavefront. In this paper presents the results of calculation of the Point Spread Function (PSF) of adaptive telescope with a circular aperture 1 m in diameter for the monochromatic wavelength (lambda) equals 0.55 micrometers . We are going to include a program imitating the operation of the wavefront sensor in numerical model, but so far assume the phase distortions in the incident wave to be well known, and the results of calculations illustrate only the limitations due to the finite number of the degree of freedom of phase corrector. We consider two types of phase correctors: the model corrector, which compensates for aberrations from tilt to coma and the segmented mirror with a hexagonal structure.

For more than two decades Lincoln Laboratory has been a major participant in adaptive-optic research and has performed seminal experiments in atmospheric compensation, including the first thermal-blooming compensation of a high-energy laser, the first compensation of a laser beam propagating from ground to space, and the first compensation using a synthetic beacon. In this paper we briefly review more than 20 years of Lincoln Laboratory work in the field.