We report the latest progress on the design, fabrication and testing of the adaptive secondary mirror to be used in the adaptive optics system to for the 6.5m upgrade to the Steward Observatory's MMT. The adaptive secondary will use electromagnetic force actuators is conjunction with a rigid reference structure to deform a thin and flexible glass facesheet. The facesheet is fabricated with figure accuracy comparable to the surface of a traditional static secondary mirror. The flexible facesheet can however, be deformed by the actuators to conjugate the changing atmospheric aberration. Capacitive position sensor are placed at each actuators and are used to rapidly measure the position of the glass facesheet relative to the rigid reference structure. These measurements are used as feedback in a servo control-loop which maintains the desired figure of the adaptive secondary facesheet. In the proposed design the mechanical interface between the facesheet and the reference structure is limited to a small hub in the center of the facesheet. Due to heat dissipation in the electromagnetic voice-coils a temperature control system is required to maintain the facesheet of the adaptive secondary near the ambient temperature of the atmosphere. We report on laboratory test of a nearly full size 60 actuator adaptive secondary prototype. We include test of actuator stroke and position accuracy, control-loop stability, and closed-loop bandwidth.

During recent years, the number and capabilities of adaptive optics systems increased considerably. Meanwhile, this observing technique has been extensively applied in various fields of astronomical research. The obtained result are numerous and illustrate impressively the stunning potential of this method which offers diffraction-limited angular resolution for ground -based telescopes. The present review summarizes the latest achievements concerning the scientific results obtained with adaptive optics.

A 941 channel, 1500 Hertz frame rate adaptive optical (AO) system has been installed and tested in the coude path of the 3.5m telescope at the USAF Research Laboratory Starfire Optical Range. This paper describes the design and measured performance of the principal components comprising this system and present sample results from the first closed-loop test of the system on stars and an artificial source simulator.

The University of Hawaii adaptive optics program has scaled its previously successful 13 elements AO system to 36 actuators and named it 'Hokupa'a', meaning 'immovable star' in Hawaiian. First light for Hokupa'a in early November of 1997, was on the Canada France Hawaii Telescope on Mauna Kea, an f/35, 3.35 meter telescope. Performance at the telescope has now been measured and compares favorably with that predicted theoretically. The extension to 36 elements has now allowed the system to give diffraction limited performance down to I band on stars as faint as 12.5 magnitude in median 0.7 arcsecond seeing on Mauna Kea. Like our previous system, extensive computer simulations were carried out to achieve the best possible match between the curvature WFS and the deformable curvature mirror.

The Max-Planck-Institutes for Astronomy and for Extraterrestrial Physics (MPE) have recently installed a laser guide star (LGS) adaptive optics (AO) system at the 3.5m telescope on Calar Alto in Spain. The AO system consists of a Shack-Hartmann sensor, a deformable mirror with 97 actuators, and a wave-front processor that allows closed loop operations of up to 1200 Hz. As a first step we closed the high order AO loop on bright natural guide stars. As a second step we closed the AO loop ALFA's design, operation, and upgrade plans.

During the nights of 18-22 March 1998, the Palomar 200 adaptive optics (AO) breathed its first breathe of life, closing lock on a natural guide star for the first time. We present telescope result obtained during the initial closed- loop tip/tilt test of the AO system at the 5m Hale Telescope on Palomar Mountain. (PHARO). Early tip/tilt correction result demonstrate a factor of approximately 1.3 improvement in peak IR irradiance for several second exposures. The facility AO and PHARO systems are scheduled for a series of engineering runs throughout 1998. Initial closure of the 241 actuator high order natural guide star adaptive loop is expected this summer.

The Chicago Adaptive Optics System has been installed on the 3.5 meter ARC telescope at Apache Point for two years. Strehl improvements of 12 are obtained at 0.85 microns in 1- 2 arcsecond seeing. The major limitation in performance is set by the telescope vibrations.

The National Solar Observatory at Sacramento Peak is developing adaptive optics (AO) for solar astronomy. We are currently implementing a low order adaptive optics system that will correct approximately 20 Zernike modes. The system design permits future expansion future expansion to about 80 Zernike modes. We are using a correlating Shack-Hartmann wavefront sensor. Since a point source is generally not available on the sun, wavefront tilts have to be derived by applying correlation techniques to images of extended objects, such as granulation, probes and sunspots.

The Multiple Mirror Telescope (MMT) is now being replaced by a single 6.5m telescope which will have an integral adaptive optics system optimized for near IR operation. We illustrate key results obtained with low-order adaptive optics at the MMT and report on progress on the major components of the system for the new telescope. We also give a brief update on the status of the telescope itself. First light for the adaptive system is expected in mid 1999.

The Canada-France-Hawaii Telescope Adaptive Optics system has now been on the sky for two years. During that time it has been installed on the telescope 13 + times with over 100 nights of scheduled observing time. It is routinely producing red and near IR images with 0.10 arcsecond full- width half-max under median seeing conditions. We present the experience we have gained using this system. We discuss installation and set up including characterization, tuning, and focusing. We consider observer training on and following use of the system. And we consider support issues.

The size of the corrected field of view of an adaptive optics system is though to be a severe limitation on the usefulness of an adaptive optics system. We show that under good seeing conditions the useful field of view is an order of magnitude larger than the field predicted by the classical isoplanatic angle. The corrected field of view is experimentally shown to increase when using spatially degraded wavefront corrections. Two methods of spatially degrading the wavefront were tested: low-order modal wavefront reconstructions and the use of a low-altitude Rayleigh wavefront reference beacon. In both cases the on- axis performance is sacrificed for an increased field of view.

In order to image faint stellar companions from the ground, high resolution adaptive optics (AO) is needed to correct atmospheric seeing. The 3.5m AO system at the Starfire Optical Range (SOR) was used to confirm theoretical models of scattered light. Halo levels were reduced by a 50 percent at 0.5 inch, which, when combined with the increased Strehl ratio, improved the gain by a factor of 23. Speckle lifetimes ranged form 20 ms at 0.1 inch to 4 ms at 0.4 inch, much longer than the 0.7 ms correction time. These results show good agreement with predictions for current technology and reveal no limitations in principle to detection of exo- solar planets using AO systems with higher speed and resolution.

The Gemini Telescopes Project is a collaboration to develop two leading edge 8 meter telescopes; Gemini North on Mauna Kea, Hawaii, and Gemini South at Cerro Pachon, Chile. These telescopes will exploit the excellent natural image quality that these telescopes will provide, a natural guide star (NGS) adaptive optics (AO) system is being developed for Gemini North.

We present preliminary results from a comparison of image estimation and recovery algorithms developed for use with advanced telescope instrumentation and adaptive optics systems. Our study will quantitatively compare the potential of these techniques to boost the resolution of imagery obtained with undersampled or low-bandwidth adaptive optics; example applications are optical observations with IR- optimized AO, AO observations in server turbulence, and AO observations with dim guidestars. We will compare the algorithms in terms of morphological and relative radiometric accuracy as well as computational efficiency. Here, we present qualitative comments on image results for two levels each of seeing, object brightness, and AO compensation/wavefront sensing.

Proposed is the novel method of dynamic nonlinear-optical correction for distortions in wide spectral band. The method is based on combining of the negative optical feedback correction and dynamic holography correction in the system, using optically addressed phase modulators.

Turbulence-induced phase aberrations cause the fringe pattern observed in an optical interferometer to change randomly. For long exposure measurements, the aperture-averaged mean-square differential phase between apertures comprising the interferometer baseline attenuates fringe visibility measurements, and thus reduces the ability to recover spatial-frequency spectrum information for the object being imaged. Application of modal atmospheric phase correction in each aperture of the interferometer reduces the degradation of visibility measurements when the turbulence-induced phase is measured with a wave front sensing device. For objects of interest in optical interferometric imaging, light levels are insufficient for adequate wave front sensing. Thus, a nearby naturally-occurring source must be used for modal compensation. Since this beacon is separated from the object being imaged, angular anisoplanatic effects are introduced which limit the benefits gained by the turbulence compensation. In this work, we quantify the interaperture correlation of modal aberrations, including finite outer scale and angular anisoplanatic effects. These quantities determine the residual phase upon correction of each aberration mode. For given levels of modal correction and relative aperture diameter Diro, we establish the maximum angular separation between the beacon and the object being imaged that corresponds to an aperture-averaged mean-square residual differential phase less than 1 rad2 . For Diro = 5, we show this limit to be approximately 3 times the isoplanatic angle, 0o . For D/ro = 30, however, the beacon offset is limited to approximately 0.2 0o Keywords: optical interferometry, atmospheric optics , adaptive optics

This paper looks at the detection limits using the new adaptive optics system on the 3.5m SOR telescope, reported by P. Ryan et al. at this conference. We discuss recent advances in our understating of the nature of high frequency wavefront structure. We analyze the various sources of high- frequency wavefront error, arising from residual adaptive optics errors, and other sources of scattered light. We characterize the AO system efficiency in the K band in terms of profile, noise level and noise time correlation. A sharp edge coronograph was used and it improved the performance as much as needed to detect brown dwarfs. Finally we give the exposure time required to detect such companions using the actual AO associated with the coronographic device, and we also point out what needs to be improved on the AO system as well as on the coronographic device in order to reach the level required for detecting exoplanets.

We present an extension of adaptive optics sky coverage calculations to include spectrometer slit power coupling. As an example, we show several slit coupling sky coverage calculations for the Gemini-North telescope. Our calculations quantify the effect on sky coverage of previous work, where we showed Strehl is not in general a good predictor of slit coupling, and provide an example of the utility of instrument-specific calculations.

We report on new results of simultaneous measurements of sodium layer column density and the absolute return flux from laser guide stars created by a monochromatic approximately 1 W CW laser, tuned to the peak of the sodium D2 hyperfine structure. The return was measured at the MMT while the sodium abundance was measured at the CFA 60 inch telescope, about 1 km away, with the Advanced Fiber Optic Echelle spectrograph. The laser frequency stability, which can greatly affect the return flux, was monitored at the same time in order to improve the measurement accuracy. After the correction for laser frequency jitter and atmospheric transmission, the absolute flux return above the atmosphere for circularly polarized light is 1.2 X 10⁶ photons s^-1 m^-2 per watt launched above the atmosphere, per unit column density, which we taken as our measured mean over the year of N(Na) equals 3.7 X 10⁹ cm^-2 at Tucson. The solidification of a final well-determined relationship between the sodium laser guide star brightness and sodium layer column density is pivotal in the design of the next generation laser guide star adaptive optics systems. We also report the measurements and analysis of the relationship between the projected beam waist of the sodium laser and the resultant spot size on the sodium layer under typical atmospheric conditions. Since wavefront measured error is proportional to spot size, and also to 1/(root) power, minimum spot size is crucial for lowest laser power requirement. By projecting the laser through diffraction limited optics of 0.5 m diameter, roughly 3 r₀, we have achieved the smallest artificial beacon yet recorded, about 0.8 arcsec.

The Lick Observatory guide star laser has provided a beacon sufficient to close the adaptive optics loop and produce corrected images during runs in 1996 and 1997. This report summarizes measurements of the wavefront quality of the outgoing beam, photoreturn signal forth sodium beacon, and radiance distribution of the guide star on the sky, and follows with an analysis of the impact of the laser on adaptive optics system performance.

A laser system to generate sodium-layer guide stars has been designed, built and delivered to the Keck Observatory in Hawaii. The system uses frequency doubled YAG lasers to pump liquid dye lasers and produces 20 W of average power. The design and performance result of this laser system are presented.

The techniques proposed in literature in order to correct the scientific object image motion using Laser Guide Star tilt measurements are briefly reviewed. Emphasis is given on comparison among the basic ideas of the various techniques. An analysis of these ideas allows to identify the limitations and error sources of different techniques, and to discuss their performances in terms of achieved tip-tilt measurement accuracy. The use of multiple laser guide star wavefront sensing schemes is considered as a way to increase the performances of the analyzed techniques.

Simple adaptive optical system are studied that can provide real-time atmospheric turbulence compensation at IR wavelengths for large ground-based astronomical telescopes. A tip-tilt mirror and a 7-element segmented mirror, a long with an iterative adaptive control method, called Far-Field Optimization, provide the turbulence correction. The primary advantage of far-field optimization is that no near-field phase measurements are required, thus eliminating the need for a wavefront sensor that is necessary in conventional adaptive optical systems. Far-field optimization may also somewhat relax the requirements for an artificial guide star. Computer simulations are used to investigate the application of far-field optimization in the presence of drifting atmospheric turbulence and as a function of the spatial granularity and the temporal bandwidth of the adaptive mirror.

The Gemini Adaptive Optics System, under construction at the Dominium Astrophysical Observatory of the National Research Council of Canada's Herzberg Institute of Astrophysics is unique among AO systems. Altair is designed with its deformable mirror (DM) conjugate to high altitude. This concept is only practical at an observatory where extraordinary measures have been taken to reduce local seeing degradation. We summarize these measures. We then describe Altair. Both the wavefront sensor foreoptics and control system are unconventional, because the guide star footprint on an altitude-conjugated DM moves as the guide star position varies. During a typical nodding sequence, where the telescope moves 10 arcseconds between exposures, this footprint moves by half an actuator and/or WFS lenslet. The advantages of altitude conjugation include increased isoplanatic patch size, which improves sky coverage, and improved uniformity of the corrected field. Although the initial installation of Altair will use natural guide stars, it will include features to use a laser guide star with minimal rework. Altitude conjugation also reduces focal anisoplanatism with laser beacons. The infrastructure of Gemini observatory provides a variety of wavefront sensor and nested control loops that together permit some unique design concepts of Altair.

The adaptive optics system for Subaru 8.2m telescope of the National Astronomical Observatory Japan has been developed for the Cassegrain ear-IR instruments, CIAO and IRCS. The system consists of a wavefront curvature sensor with 36 subaperture photon-counting avalanche photodiode modules and a bimorph deformable mirror with 36 electrodes. The expected Strehl ratio at K band exceeds 0.4 for objects that are located close enough to a bright guide star as faint as R equals 16 mag at the median seeing of 0.45 arcsec at Mauna Kea. The system will be in operation in 1999 as a natural guide star system, and will eventually be upgraded to a laser guide star system in cooperating an IR wavefront tilt sensor to provide nearly full sky. The construction of this common use system to Subaru telescope is now underway in our laboratory in Tokyo. Prior to starting the fabrication of this common use system, a full size prototype system was constructed and tested with the 1.6 m IR telescope at our observatory in Tokyo. This system has the identical optical design, deformable mirror, loop control computer to those for the Subaru system, while the wavefront sensing detectors were less-sensitive analog APDs. We succeeded in getting closed loop images of stars in K band with diffraction limited core. The Strehl ratio was around 0.5 and the factor of improvement was about 20 at K-band under the average seeing of 2 arcsec during the observation. The loop sped of the system was 2 K corrections per second.

NAOS is the adaptive optics system to be installed at one of the Nasmyth foci of the very large telescope (VLT). It will provide compensated image to the high angular resolution IR spectro-imaging camera which covers the 1-5 micrometers spectral bands. our French consortium is the sub-contractor of ESO for the design, manufacturing, integration and test of NAOS. For bright sources, the specification is to reach 70 percent Strehl ratio under average seeing conditions. Two wavefront sensors, one in the visible spectral range and one in the near IR spectral range, will equip the adaptive optics system. We foresee to obtained the first light at the VLT unit telescope 1 in mid-2000.

The laser guide star adaptive optics system currently being developed for the Keck 2 telescope consists of several major subsystems: the optical bench, wavefront control, user interface and supervisory control, and the laser system. The paper describes the design and implementation of the wavefront control subsystem that controls a 349 actuator deformable mirror for high order correction and tip-tilt mirrors for stabilizing the image and laser positions.

The astronomical applicability of laser beacons depends critically on the ability to track on a faint object to less than the diffraction limit of the telescope. Detailed measurements taken at Apache Point Observatory and a search of published of the literature shows that measured power spectrum are often very different from the theoretical predictions of Greenwood and Tyler. This has important consequences and it is possible that the formula usually used to design the tip/tilt servos are in error, often by large factors, and that much wider band-widths are required both for the tip/tilt and higher order correction.

We describe different works conducing to the adaptive optics system for the TIM 6.5m telescope. We show turbulence profiles result at our San Pedro Martir Observatory in Baja using the Generalized SCIDAR. We can conclude that the turbulence conditions in this site are comparable to the major observatories in the world. From these results and taken in account curvature AO simulations it is possible to predict the performances in limiting magnitude and sky coverage of different AO systems and telescopes in our observatory. We can also define the degree of the AO system for the TIM 6.5m telescope. We made a short description of our LOLA tip-tilt corrector system and the GUIELOA 19 elements curvature AO system. The calculation of the optics quality for the TIM 6.5m is briefly mentioned. Studies about the influence of the finite outerscale on the optical quality of AO corrected images are described.

We discuss the current developments and the perspective performances of adaptive secondary mirrors for high order adaptive a correction on large ground based telescopes. The development of the basic techniques involved a large collaborative effort of public research Institutes and of private companies is now essentially complete. The next crucial step will be the construction of an adaptive secondary mirror for the 6.5 m MMT. Problems such as the fabrication of very thin mirrors, the low cost implementation of fast position sensors, of efficient and compact electromagnetic actuators, of the control and communication electronics, of the actuator control system, of the thermal control and of the mechanical layout can be considered as solved, in some cases with more than one viable solution. To verify performances at system level two complete prototypes have been built and tested, one at ThermoTrex and the other at Arcetri. The two prototypes adopt the same basic approach concerning actuators, sensor and support of the thin mirror, but differ in a number of aspects such as the material of the rigid back plate used as reference for the thin mirror, the number and surface density of the actuators, the solution adopted for the removal of the heat, and the design of the electronics. We discuss how the results obtained by of the two prototypes and by numerical simulations will guide the design of full size adaptive secondary units.

We report the latest progress on the design, fabrication and testing of the adaptive secondary mirror to be used in the adaptive optics system to for the 6.5m upgrade to the Steward Observatory's MMT. The adaptive secondary will use electromagnetic force actuators in conjunction with a rigid reference structure to deformed a thin and flexible glass facesheet. The facesheet is fabricate with figure accuracy comparable to the surface of a traditional static secondary mirror. The flexible facesheet can however, be deformed by the actuators to conjugate the changing atmospheric aberration. Capacitive position sensors are placed at each actuator and are used to rapidly measure the position of the glass facesheet relative to the rigid reference structure. These measurements are used as feedback in a servo control- loop which maintains the desired figure of the adaptive secondary facesheet. In the proposed design the mechanical interface between the facesheet and the reference structure is limited to a small hub in the center of the facesheet. Due to heat dissipation in the electromagnetic voice-coils a temperature control system is required to maintain the facesheet of the adaptive secondary near the ambient temperature of the atmosphere. We report on laboratory test of a nearly full size 60 actuator adaptive secondary prototype. We include test of actuator stroke and position accuracy, control-loop stability, and closed-loop bandwidth.

We present the result of electrical and optical measurements performed on a reduced size adaptive secondary prototype named P30. The design of this concave deformable mirror consists of a thin deformable glass shell whose figure is controlled by electromagnetic actuators and capacitive position senors. Static measurements of the mirror optical figure, performed with a commercial interferometer, have provided the calibration of the internal position sensor. Dynamic test were performed to experimentally derive the mechanical response of the mirror to the electromagnetic actuators in order to design the mirror closed loop control law. The test, although performed on a reduced scale, are representative of the complexity and capabilities of the full size mirror. In fact, all the key-elements of the full size mirror, i.e. central supporting membrane, actuator spacing closed loop control of the device, have been implemented on the prototype.

In this paper we present a status report of our liquid crystal (LC) adaptive optics program. Liquid crystal have been proposed to be used for adaptive optics components by several authors. We are actively pursuing a two prong approach, on one side we have built a closed loop test bed based on a 127 element nematic LC device; on the other side we are actively involved in research on the ferroelectric side of the LC devices. In this paper we present preliminary results from our testbed. The essential elements of this testbed are a Shack-Hartman based wavefront sensors with an analog reconstructor, and the 127 nematic elements from Meadowlark Optics. Some of the basic experimental parameters will be reviewed, and result will be discussed. Furthermore we will present the status of our activity in the field of ferroelectric LC. Several prototypes are under testing to establish their capabilities as AO components.

Nematic liquid crystals have been sued as phase retarders in adaptive optics system for more than 10 years. Liquid crystal spatial light modulators (SLMs) can be very useful both as controlled disturbances for characterizing adaptive optics systems, and as wavefront correctors. Of the available options, SLMs have the advantages of low cost, reliability, compactness, low power, ease of controllability, optical transmissivity, and simple aperture scalability. Moreover, these devices are very non-linear and normally have quite different response times in the increasing versus the decreasing phase direction. The current limitation for using these devices in atmospheric adaptive systems is their relatively slow response times, not only for small phase steps in the excitation direction, but for all steps in the relaxation direction. The response time in both directions can be significantly improved by exploiting two important properties of certain nematic liquid crystals: the Dual-Frequency Effect and the Transient Nematic Effect. This paper presents the details of custom- built electronic hardware that takes advantage of these two effects to implement (lambda) /10 phase accuracy with millisecond settling time for both increasing and decreasing phase shifts.

This paper presents recent result for aberration correction and beam steering experiments using polysilicon surface micromachined piston micromirror arrays. Microfabricated deformable mirrors offer a substantial cost reduction for adaptive optic systems. In addition to the reduced mirror cost, microfabricated mirrors typically require low control voltages, thus eliminating high voltage amplifiers. The greatly reduced cost per channel of adaptive optic systems employing microfabricated deformable mirrors promise high order aberration correction at low cost. Arrays of piston micromirrors with 128 active elements were tested. Mirror elements are on a 203 micrometers 12 by 12 square grid. The overall array size is 2.4 mm square. The arrays were fabricated in the commercially available DARPA supported MUMPs surface micromachining foundry process. The cost per mirror array in this prototyping process is less than 200 dollars. Experimental results are presented for a hybrid correcting element comprised of a lenslet array and piston micromirror array, and for a piston micromirror array only. Also presented is a novel digital deflection micromirror which requires no digital to analog converters, further reducing the cost of adaptive optics system.

This paper describes the design, fabrication, modeling, surface characterization, and simulation of advanced surface micromachined micromirror devices that are optimized for adaptive optics applications. Design considerations and fabrication capabilities are presented. Simulation of adaptive optics performance of unique Flexure-Beam and Axial-Rotation Micromirror devices is performed for many common aberrations. These devices are fabricated in the state-of-the-art four-level planarized polysilicon process available at Sandia National Laboratories known as the Sandia Ultra-planar Multi-level planarized MEMS Technology. This enabling process permits the development of micromirror devices with near-ideal characteristics that have previously been unrealizable in standard three-layer polysilicon processes. This paper describes such characteristics as elevated address electrodes, array wiring techniques, planarized mirror surfaces using chemical mechanical polishing, unique post-process metallization, and the best active surface area to date.

Multi-segment liquid crystal spatial light modulators have received much attention recently for use as high-precision wave-front control devices for use in astronomical and non- astronomical applications. They act much like piston only segmented deformable mirrors. In this paper we investigate the use of these devices in conjunction with a Shack- Hartmann wavefront sensor. In this paper we consider a zonal algorithm in order to take advantage of high speed matrix multiply hardware which we have in hand.

We describe and evaluate the performance of a wavefront sensor based on curvature sensing which can be used to detect static aberrations given an extended reference source. The description includes a full mathematical treatment of the sensor signal, as well as how this signal is relate to the Laplacian of the wavefront. Evaluation of the technique is performed with computer simulations. A Monte-Carlo simulation is utilized to evaluate the performance of the technique in the presence of noise. The sensor was found to provide accurate measurement of the wavefront coefficients on high-contrast extended objects. It behaves well in the presence of a field stop, and in the presence of additive Gaussian noise.

We compare phase diversity and curvature wavefront sensing. Besides having completely different reconstruction algorithms, the two methods measure data in different domains: phase diversity very near to the focal plane, and curvature wavefront sensing far from the focal plane in quasi-pupil planes, which enable real-time computation of the wavefront using analog techniques. By using information- theoretic lower bounds, we show that the price of measuring far from the focal lane is an increased error in estimating the phase. Curvature wavefront sensing is currently operating in the field, but phase diversity should produce superior estimates as real-time computing develops.

We present a prototype of a Foucault-Like Wavefront Sensor (FLWS) that will be implemented in the Adaptive Optics Systems of 3.5m TNG and is of the proposed choices for the 2 X 8.4m LBT. The FLWS has some interesting features for Adaptive Optics applications: it has a linear relationship between measured signals and wavefront derivatives like the Shack-Hartmann sensor, it allows to change on the fly the gain used for wavefront measurements like the Curvature Sensor, and finally it allow to modify easily the wavefront aberration spatial sampling. The developed prototype of the FLWS is described and first laboratory results are presented. Measurements of static wavefront aberrations performed by the FLWS and a Wyko interferometer are compared to test the accuracy of the sensor, giving an error of a few tens of nm in terms of reconstructed wavefront rms, when aberrations of typical atmospheric turbulence strength are induced.

We show with simulation experiments that closed-loop phase- diversity can be used without numerical guard-bands for wavefront sensing of low-order wavefronts from extended objects using broad-band filters. This may allow real-time correction at high bandwidth for certain applications. We also present a proper maximum likelihood treatment of Shack- Hartmann data, which includes an imaging model to extract curvature information from the lenslet images. We demonstrate by simple simulations that this approach should allow higher-order wavefront information to be extracted than with traditional Shack-Hartmann wavefront sensing for a given number of lenslets.

A new wavefront sensors design is presented which allows reconstruction of phase without the insensitivity to waffle mode which has previously plagued the Fried Geometry. This new system senses and reconstructs waffle, with less total computational burden, and little change in noise propagation, than previous systems.

The contradiction inherent in high temporal bandwidth adaptive optics wavefront sensing at low-light-levels (LLL) has driven many researchers to consider the use of high bandwidth high quantum efficiency (QE) CCD cameras with the lowest possible readout noise levels. Unfortunately, the performance of these relatively expensive and low production volume devices in the photon counting regime is inevitably limited by readout noise, no matter how arbitrarily close to zero that specification may be reduced. Our alternative approach is to optically couple a new and relatively inexpensive Ultra Blue Gen III image intensifier to an also relatively inexpensive high bandwidth CCD camera with only moderate QE and high rad noise. The result is a high bandwidth broad spectral response image intensifier with a gain of 55,000 at 560 nm. Use of an appropriately selected lenslet array together with coupling optics generates 16 X 16 Shack-Hartmann type subapertures on the image intensifier photocathode, which is imaged onto the fast CCD camera. An integral A/D converter in the camera sends the image data pixel by pixel to a computer data acquisition system for analysis, storage and display. Timing signals are used to decode which pixel is being rad out and the wavefront is calculated in an analog fashion using a least square fit to both x and y tilt data for all wavefront sensor subapertures. Finally, we present system level performance comparisons of these new concept wavefront sensors versus the more standard low noise CCD camera based designs in the low-light-level limit.

We discuss astronomical observing requirement sand their implementation using sodium-layer laser guide star adaptive optics. Specific issues requiring implementation include the ability to place the astronomical object at different locations within the field of view; reliable subtraction of Rayleigh-scattered light; efficient focusing; and stable ;point-spread-function characterization.

Final design details are given for UnISIS, the University of Illinois Seeing Improvement System. The principle components include a 50 Watt Excimer laser working at 351 nm which produces a pulsed Rayleigh laser guide star at 333 Hz and 18 km altitude, a 177-actuator deformable mirror, an atmospheric dispersion correction system, and a science camera configuration with both visual and near-IR cameras that can be used simultaneously. Two high-speed CCD cameras are used to feed dual quad C40 DSP-based reconstructor to close the feedback loop with the deformable mirror. The adaptive optics system is situated on a large optics table that rests above the old Coude spectrograph of the Mt. Wilson 2.5-m telescope and the Excimer laser is located in a basement Coude room.

The Mauna Kea Laser Guide Star Technical Working Group was established to address related to the coordination and use of laser beacons for adaptive optics at the astronomical observatories located on the summit of Mauna Kea, Hawaii. This paper describes the issues that are being addressed by the group and the policies that have been adopted for Mauna Kea.

The design and characteristics of an 8 watt diode laser pumped version of the sum frequency laser designed for astronomy is described together with an outline of field test undertaken on the NOAO Vacuum Telescope Tower at Sac Peak. Optical pumping effects were shown to increase the return by a factor of 2.5 subarcsecond beacons were produces with a brightness equal to a 9.2 magnitude star with 1.2 watts incident at the sodium layer under sodium abundance column density of 5 X 10⁹ atoms/cm².

Control laws for an adaptive optics system for the Palomar Mountain Hale Telescope are described. These are derived using a linear matrix model of the optics, which gives the Hartman-sensor centroids and the science camera wavefront as functions of deformable mirror (DM) commands and atmospheric phase. The matrices defining this system can be directly measured from the optics, some at finer spatial resolution than the wavefront sensor can resolve. A minimum-wavefront compensator feeds back both the wavefront sensor centroids and previous DM commands, using the atmospheric covariance to smooth the response at the finer spatial scale. The DM command feedback provides direct observability of waffle, piston and other modes unobservable in the wavefront sensor. Compensator gains can be updated using Kalman filtering techniques to track the evolution of the atmospheric covariance matrix.

We describe the current activity at ESO on Curvature Adaptive Optics Systems. The strategy, aimed at implementing several system via a cloning process, will allow faster production time,lower costs and better maintenance. Curvature systems fulfill the performance requirements for NIR spectroscopy and interferometry at the VLT, for which AO is requested. This development and the systems described are complementary to the foreseen ESO high NIR Strehl Ratio Ao system, NAOS in combination with the camera CONICA.

It is foreseen that the Gran Telescopio Canarias will include an Adaptive Optics (AO) system to provide diffraction-limited images in the near-IR. The preliminary scientific requirements are presented here and are used to determine the basic system specifications. The sky coverage is determined for the cases of natural and laser guide stars. The results of conjugating the deformable mirror to the entrance pupil and to the mean turbulence height are compared. The Gran Telescopio Canarias will employ a segmented primary mirror. This will introduce discontinuous wavefront errors due to possible effects of these errors on the AO performance. The resulting requirements for these errors are compared with those determined by the error budget for long-exposure image quality.

We will review the status of the natural/laser guide star adaptive optics facility that is being constructed for the Keck II telescope.

A description is given for the geometry of the first generation adaptive-optical (AO) system being developed for the 6.5m single-mirror telescope that will replace the Multiple Mirror Telescope on Mt. Hopkins. The AO system consists of an adaptive secondary mirror with 330-actuators and a wave front senor composed of a 13 by 13 square subaperture array. The deformable mirror actuator vibrational modes were determined from a finite element model of the adaptive-secondary mirror. Numerical simulations in based on the finite element results were carried out to determine the expected performance of the system. Two reconstruction algorithms were compared - a least squares reconstructor and a modal equalization technique developed for this AO geometry. Strehl ratios are reported for the two algorithms for various guide star magnitudes and number of corrected wave front modes.

The design of adaptive optics system requires astrophysically driven requirements and specifications. We present here the study performed to help specifying and designing the ESO Nasmyth Adaptive Optics System of the VLT.

The 8-meter Gemini telescope's adaptive optics (AO) module, Altair, is 'transparent' in that it does not change the focal ratio, being f/16 in and f/16 out; it has the same focal position as the bare telescope, with insignificant change in the exit pupil. However, Altair has a flat focal surface, unlike other AO designs which have focal surfaces curved more than the focal surface of the bare telescope and in the opposite direction. An unusual requirement for Altair is that the atmospheric layer 6.5 km above the telescope should be imaged onto the deformable mirror. Other requirements are minimization of distortion in the wavefront sensor module for both the imaging of the deformable mirror onto the lenslet array and for the reimaging of the approximately 230 lenslets' images onto a CCD, for a natural guide star, and also for a Sodium laser guide star ranging in object distance form 85 km to 156 km. The separation of natural and laser star beams is done with minimum light loss by passing the in-focus natural star image through a pinhole which is smaller than the shadow of the secondary mirror of the telescope in the out-of-focus laser beam which is reflected by the tilted pinhole mirror.

Post processing of data obtained from the European Southern Observatory's ADONIS system is presented here. A physically constrained iterative deconvolution algorithm is applied to two distinctly different objects demonstrating both object and point spread function recovery. Surface features are found in thermal imaging of Io which compare well with Galileo measurements of vulcanism. Anisoplanatic behavior of the ADONIS system is studied with wide field imaging of the cluster NGC1850.

Phase diversity is a technique for obtaining estimates of both the object and the phase, by exploiting the simultaneous collection of two short-exposure optical images, one of which has been formed by further blurring regularized variant of the Gauss-Newton optimization method for phase diversity-based estimated when a Gaussian likelihood fit-to-data criterion is applied. Simulation studies are provided to demonstrate that the method is remarkably robust and numerically efficient.

This paper addresses post processing for resolution enhancement of sequences of short exposure adaptive optics (AO) images of space objects. The unknown residual blur is removed using Bayesian maximum a posteriori blind image restoration techniques. In the problem formulation, both the true image and the unknown blur psf's are represented by the flexible generalized Gaussian Markov random field (GGMRF) model. The GGMRF probability density function provides a natural mechanism for expressing available prior information about the image and blur. Incorporating such prior knowledge in the deconvolution optimization is crucial for the success of blind restoration algorithms. For example, space objects often contain sharp edge boundaries and geometric structures, while the residual blur psf in the corresponding partially corrected AO image is spectrally band limited, and exhibits while the residual blur psf in the corresponding partially corrected AO image is spectrally band limited, and exhibits smoothed, random , texture-like features on a peaked central core. By properly choosing parameters, GGMRF models can accurately represent both the blur psf and the object, and serve to regularize the deconvolution problem. These two GGMRF models also serve as discriminator functions to separate blur and object in the solution. Algorithm performance is demonstrated with examples from synthetic AO images. Results indicate significant resolution enhancement when applied to partially corrected AO images. An efficient computational algorithm is described.

Deconvolution from wavefront sensing (DWFS) is an image reconstruction technique that can be used as an alternative to phase diversity or speckle reconstruction techniques for observations of extended objects. A correlating Shack- Hartmann wavefront sensor was used to measure the instantaneous wavefront. The wavefront information was then used to deconvolve simultaneously taken solar images. The main objective of this experiment was to validate the correlating Shack-Hartmann wavefront sensor approach for solar adaptive optics. Several series of short exposure images and simultaneous wavefront sensor data were collected at the Vacuum Tower Telescope at Sacramento Peak. We produced a series of reconstructions of solar features, including granulation, pores and sunspots. We show that consistent result can be achieved using DWFS. Reconstructed images of the same object but observed at slightly different times, i.e. different atmospheric realizations, are nearly identical.

Multiple guide stars are studied as a mean so of improving the performance of laser guide star (LGS) adaptive optical (AO) systems by reducing the effects of anisoplanatism. Previous work has shown that combining one sodium LGS and one Rayleigh LGS reduces tilt anisoplanatism for some turbulence profiles. In this paper, we investigate two scenarios of multiple guide star configurations. In the first scenario, AO correction based on three or more sodium beacons is studied. The second scenario investigates the use of a single sodium beacon in conjunction with a single Rayleigh beacon. For each of these case, two turbulence profiles, two deformable mirror conjugate altitudes, numerous tip-tilt guide star offsets, and two WFS subaperture array sizes are investigated. The best results of the two scenarios are determined and compared in terms of Strehl Ratio and residual tilt.

In addition to natural and laser guide stars for adaptive optics, it is proposed to use radio-created guide stars or fringes. Heating by intense radio beams either modulators sodium lamp illumination, or creates and modifies plasma in different altitudes. The plasma relaxes by artificial air glow, which is concentrated in few lines, mainly nitrogen and oxygen. Fringes between intense radio beams create plasma fringes, also visible from the telescope. Ionospheric heating was also shown to create patchy artificial aurora. In all cases, the multiplicity of sources or fringes allows multi-conjugate adaptive optics of wide fields of view, and shows promise for long baseline optical interferometry.

The generation of sodium guide stars through laser excitation imposes very specific requirements on the laser. It has been argued that the most effective sodium guide star laser would operate CW with a mode spacing of less than the 10 MHz Doppler broadening profile of the mesospheric sodium. However, achieving high power in free-space optics with these requirements becomes very difficult because of the limited active gain volume and exceedingly long cavity requirement. The clear solution to these impediments is the development of a fiber raman laser (FRL) to shift the output of a solid state laser to a harmonic of the 589 nm sodium transition. The specific wavelength and bandwidth are selected from the large Raman gain spectrum by application- specific Distributed Bragg Reflectors. The guided-wave nature of the FRL allows for very large interaction volumes, while the long fiber length allows for narrowly-spaced modes which easily fill the sodium D₂ continuum. This paper reviews the theory and design of the Fiber Raman Guidestar Laser being developed for Astronomical Adaptive Optical systems.

VLT is taken as a case study for the tilt retrieving scheme from a laser guide star using the auxiliary telescopes techniques. Assuming that the auxiliary apertures can be moved anywhere on the set of rails surrounding the four VLT units, the corresponding patch on the sky useful to find out a suitable natural guide star for tracking purposes is evaluated. This is accomplished taking into account both the thickness of the sodium layer and the isoplanatic patch size that one can reasonably expected under good seeing conditions at the telescope site. It is shown that the useful patch area to be looked for NGSs rotates and changes in shape in the sky during the observation of a single target. The sky coverages estimated for two cases are given, this allows to trace some preliminary conclusions on the feasibility of such tilt retrieving scheme.

Kolmogorov turbulence has a temporal behavior best described by fractional Brownian motion (FBM) with a self-similarity parameter of H equals 5/6. Non-Kolmogorov behavior, which has also been observed and reported, is characterized by H other than 5/6. Although FBM has strong inter-sample correlations, the signals arising in an adaptive optics control loop are not FBM, have weaker inter-sample correlation,and exhibit quite different characteristics with regard to predictability. One conclusion is that for greatest advantage, prediction within the loop should be introduced at the point where signals represent the reconstructed wavefront, i.e. mirror drive signals, and not at the level of the wavefront sensor output or wavefront error signals.

The ability to generate long sequences of spatially and temporally correlated phase and slope screens is critical to simulating adaptive optics system. Covariance based methods exist for generating short sequences of phase and slope screens which theoretically could produce the long sequences needed. However, implementing these methods for large sequences of frames is impractical with today's computer technology. FFT-based methods may be used to generate unlimited-length sequence which are periodic. This paper present a method for generating unlimited length sequences which are not periodic using optimal estimators. This method does not yield perfect spatial and temporal statistics for screens separated by large numbers of frames in time. However, the short term statistics for any frame are accurate, while the errors in long term statistics are small.

The Starfire Optical Range (SOR) division of the Air Force Research Laboratory operates a 3.5m aperture telescope system near Albuquerque, New Mexico. The SOR conducts advanced research into adaptive optics and the imaging of space objects. September of 1997 saw the 'first light' for a 941 actuator adaptive optics system for this 3.5m telescope.

We continue our discussion of various mathematical approaches for matrix estimators for zonal adaptive optics systems which are designed and optimized for different situations.

In recent years several methods have been presented for optimizing closed-loop adaptive-optical (AO) wave-front re- construction algorithms. These algorithms, which can significantly improve the performance of AO systems, compute the reconstruction matrix using measured atmospheric statistics. Since atmospheric conditions vary on time scales of minutes, it becomes necessary to constantly update the reconstruction so that it adjusts to the changing atmospheric statistics. This paper presents a method for adaptively optimizing the reconstructor of a closed-loop AO system in real time. The method relies on recursive least square techniques to track the temporal and spatial correlations of the turbulent wave-front. The performance of this method is examined for a sample scenario in which the AO control algorithm attempts to compensate for signal processing latency by reconstructing the future value of the wave-front from a combination of past and current wave-front sensor measurements. For this case, the adaptive reconstruction algorithm yields Strehl ratios within a few percent of those obtained by an optimal reconstructor derived from a priori knowledge of the strength of the turbulence and the velocity of the wind. This level of performance can be a dramatic improvement over the Strehls achievable with a conventional least squares reconstructor.

Previous work has shown that atmospheric distortions are predictable. We present preliminary results which show that introducing predictors into the control loop of an adaptive optics system with a zero hysteresis segmented mirror improves systems performance.

The recent concept of an adaptive secondary mirror controlled by a large number of force actuators has been explored through the realization of a thirty actuators prototype. The design and implementation of the local position control electronics and real time software is presented. Both centralized and de-centralized control approaches are considered, and for the experimental implementation the de-centralized solution is adopted. The design is based on a digital approach. Mirror actuators are controlled by an array of DSPs. Each DSP controls two actuators: the position measurement are obtained from capacitive sensors co-located with the actuators and a local closed loop position control is implemented on the DSP to drive the voice coils. In the prototype unit, a bi- direcitonal fast serial line is used to exchange data with the wavefront computer. The communication link can be used to download program code and data to the DSPs thus allowing the greatest flexibility in the definition of algorithms and working parameters. The same link is used during mirror operation to issue position commands to the deformable mirror and to get back diagnostic data.

During 1996 and 1997 more than 20 nights observing time have been used by, or allocated to, the CHARPA group at Georgia State University using the Mount Wilson Institute Adaptive Optics mounted on the Hooker 100 inch telescope on Mt. Wilson. Several scientific programs are being pursued including: differential photometry of binary stars; a search for faint companions of local solar type stars; attempts to image dust shells around YSOs; and experiments involving the combination of non-redundant aperture masking interferometry and adaptive optics. We have learned, and continue to learn, a great deal about the problems associated with, and methods of calibration of, adaptive optics images, especially in the area of accurate photometric measurements. So far, more than 30 binary systems have been measured in multiple filters and several previously unknown faint companions to local stars have been identified.

The intense stellar radiation of deeply embedded young massive stars leads to the formation of very compact ionized regions. Adaptive optics observations of such objects using ESO's ADONIS system yielded diffraction-limited images at 2.2 and 3.6 micrometers . Supplementary measurements were acquired at mid-IR and radio wavelengths. As a main result, it turned out that the observationally defined class of UCHIIs comprises sources of different nature: ionized dense gas clumps close to massive stars, externally ionized circumstellar disks and photo-evaporating disks around hot stars. In any case, the reservoir provided by the molecule matter prolongs the lifetime of an UCHII over that of a classical HII region which explains the high frequency of galactic UCHIIs. The investigations yielded new insights on the birth of massive stars and the interaction with the ambient matter of the parent molecular cloud which helps to establish a more accurate theory of the formation of massive stars and to describe properly their impact on the stimulation/inhibition of star formation in general.

The University of Hawaii AO group has been actively carrying out astronomical AO observations for the last four years. The UHAO group and out collaborators have utilized the curvature AO system to obtain diffraction-limited images of asteroids, planets, moons, protoplanetary disks, young stars, young star clusters, planetary nebulae, black holes, galaxies and quasars. The current scientific capabilities of the new 36-actuator Hokupa'a AO curvature system will be briefly reviewed. Four key astronomical situations that are excellent for AO observations will be discussed. Examples of scientific observational techniques will be highlighted with actual AO astronomical results.

High angular resolution adaptive optics data have been obtained on (eta) Carinae and Homunculus nebula, with the ESO ADONIS system. In the broad band filters, J, H and K, the central source, (eta) Car was allowed to saturate in order to maximize the signal for the extended Homunculus nebula. Unsaturated images were also obtained in a narrow continuum filter. The purpose of these observations was to explore the 3D shape of the Homunculus through high spatial resolution imaging polarimetry. The polarization of resolved knots and filaments, will allow their position along the line of sight to be determined when combined with scattering models. Here the observational details, data reduction and preliminary results are presented.

We present a method to accurately find the position and photometry of each source in a stellar field blurred by the adaptive optics corrected point spread function (AO PSF). The method takes into account the specificities of the object as well as the specificities of the AO PSF. The algorithm is based on the minimization in the Fourier domain of a regularized least square objective function using the Levenberg-Marquardt method. Astrometric and photometric data of each source are retrieved with an excellent accuracy and super-resolution is achieved. The performance of the method in terms of accuracy is assessed on simulated fields. Then, we present results obtained on images of the Galactic Center, a very crowded field on which the method shows remarkable robustness and stability.

Following extensive development effort, approximately a dozen adaptive optics facilities are now available for research in astronomy, and a similar number is nearing competition or in advanced planning. The scientific productivity, measured by research papers, is rapidly increasing. From a survey of published research and a review of research provisionally discuss the contribution of natural guide star adaptive optics to astronomy. The most active research topics for adaptive optics astronomy have been in solar system studies and in the observation of young stars and star forming regions. The benefit of adaptive optics most prominently exercised in these observations has been high resolution imagery, and the most common area of concern is the point spread function. The scientific success supports the position that adaptive optics will son be required for large telescopes to remain competitive in certain research areas. At the same time, most areas of astronomy research remain untouched by adaptive optics techniques.

We summarize the history and our current and planned adaptive efforts using the telescopes at Yerkes Observatory.

It has been demonstrated both with computer simulation and field experiments that low order adaptive optics can improve the signal to noise ration in subsequent computer post- processing algorithms, especially when viewing an extended object. In this presentation we show open and closed loop images of a meteorological satellite taken with a low order adaptive optics instrument package mounted on the side of the Air Force's 3.5 meter telescope. The images are subsequently post-processed using speckle algorithms. It is shown that the boost in post-processing signal to noise ratio, provided by the low order adaptive optics, closely corresponds to computer simulation results.

The Keck AO system is nearing completion and is nearly ready for system integration in the optics laboratory at Keck Headquarters. because of the ambitious scientific objectives of this project, the AO system is quite complex in design. The optics bench alone has 33 motion control axes, and potentially 6 image planes. Calibration will play a major role in making the AO system function as planned. We have identified approximately 35 separate calibration procedures that will be needed to be performed during the integration phase in the laboratory. Many of these procedures are quite straightforward. Other calibrations are quite involved and are mini 'research projects' in themselves. It is this last category that will be discussed in this paper.

The adaptive optics modules for the Italian Telescopio Nazionale Galileo, namely AdOptTNG, is ready to be mounted at the telescope. First light will take place with the tip- tilt loop working, about which further laboratory test has been made during the last month. Also, the speckle camera with the real time autocorrelation feature has been completely finished and tested. The tracking system which allow to use moving references has been completed. For what regards the higher order loop the contracts about the DM and the CCD controller have been successfully concluded. Moreover, the wavefront sensing unit has been in large part built and tested in the laboratory. Details of the integration in the Nasmith Interface carrying the CCD scientific camera and the IR Imager are given together with a thorough description of the internal acceptance test of the optomechanical module comprising more than 20 motorized axes and several functioning modes.

The sodium laser guide star adaptive optics system ALFA has been constructed at the Calar Alto 3.5m telescope. Following the first detection of the laser beacon on the wavefront sensor in 1997 the system is now being optimized for best performance. In this contribution we discuss the current status of the launch beam and the planned improvements and upgrades. We report on the performance level achieved when it is used with the adaptive optics system, and relate various aspects of our experience during operation of the system. We have begun to produce scientific result and mention two of these.

Described are the methods of numerical design and simulation of observational telescope, using the thin dynamic hologram as the corrector for image distortions, imposed by the primary mirror surface defects.

The combination of adaptive optics and chronographic techniques providing at the same time high angular resolution and high dynamic range is a powerful tool to study the close environment of bright objects in different fields of astrophysics. A stellar coronography dedicated to the ESO adaptive optics system ADONIS has been designed and manufactured by Observatoire de Grenoble and is now offered to the astronomical community. We present here the design and performance analysis of this new coronograph and illustrate its impact for a few astrophysical programs.

The use of a Laser Guide Star (LGS) in an Adaptive Optics system in order to increase the sky coverage, leads to a degradation of correction performances caused by the focus anisoplanatism, or cone effect, the LGS upward tilt indetermination and the defocusing produced by the non-constant altitude of the LGS. If the Adaptive Optics system has the typical configuration of a Shack-Hartmann wavefront sensor and a piezo-stack deformable mirror, a possible partial solution to solve the problem is to add a low order curvature wavefront sensor able to look at an off-axis a natural guide star close enough to the observed object. In fact the very low order modes (tip-tilt, defocus and astigmatism) are the most degraded by the LGS use but they have also the largest isoplanatic patch especially in the near infrared (H,K bands) wavelengths. With this configuration the Shack-Hartmann sensor provides the coefficients for the higher order modes while the curvature sensor provide the coefficients for the lower order ones. In this way the tilt and conical anisoplanatisms are partially reduced and there is also a continuos monitoring of the sodium layer altitude. Furthermore another aspect is the comparison of the performances at very faint magnitudes between a Shack-Hartmann and a low order curvature sensor in order to check if the reduced pupil sampling provides a better correction. In this framework a simulation work has been developed to study the coupling performances of a curvature sensor with a piezo-stack mirror; the code has been written in IDL. Keywords: adaptive optics, laser guide stars, numerical simulations, curvature wavefront sensors, deformable mirrors

We describe the design of the ESO Adaptive Optics (AO) systems for the very large telescope interferometer (VLTI). We consider hereafter both the tip-tilt only corrections and the high order systems. The high order AO systems are designed for K-band operation on the Unit Telescopes (UT). The K-band UT beams will be combined with the 1.8m Auxiliary Telescopes (AT) operating with tip-tilt correction only, via ESO's Strap system. The UT-AO system will be hosted in the Coude' laboratory, with the deformable mirror inserted at the M8 location of the optical train. The wavefront sensor retains the option to be either in the Coude' lab, before the delay lines, or at the end of the beam combining path in the Interferometry laboratory, depending on the instrument attached and its use.

We have analytically modeled the performances of a natural guide star and laser guide star adaptive optics system. The performance in terms of Strehl radio in J, H and K bands are presented for two atmospheric models. We use this model to derive values of sky coverage for both systems following two methods. The first one is statistical and uses a model of our Galaxy to give the stellar densities and the AO model to compute the off-axis performances. The second model uses a cross correlation of catalogues of astrophysical objects and reference stars. This gives the number of objects of a given kind that can be observed with a given Strehl. We have also modeled the PSF of a laser guide star system using the modal decomposition of the cone effect and the effect of tilt isoplanatism. It can be seen that the PSF from an LGS system differs significantly from that of a conventional AO system.

We describe in this paper the development of a simulation tool for modeling any kind of adaptive optic system with laser guide star within the framework of the European network Laser Guide Star for 8-m Class Telescopes. This is intended to support investigations about laser guide stars problems line cone effect and tip-tilt determination. The different kinds of software modules to be developed are listed and the libraries already distributed among the network are described: phase screen generation, downward propagation, wavefront sensing and reconstruction. The future studies for solving the cone effect are also presented.

Generation of sodium guide stars for adaptive optics requires very precise control of the frequency and bandwidth of the laser to maximize the brightness of the generated guide star. The ruggedness, efficiency and ease of use of a solid state system has great potential for improving the reliability and power of the laser guide star over the dye laser system currently used. The dearth of solid state transitions at the precise wavelength required for exciting resonance scattering in sodium drives us toward Raman shifting to downshift a nearby solid-state transition line tuned to work with the Raman-shifting material. The system being developed for the 6.5 meter multiple mirror telescope (MMT) takes two approaches to creating the sodium guide star: one uses YGAG to maximize the Raman-shifted output at the sodium D₂ resonance. The second approach is to thermally tune the output of YAG to reach the appropriate wavelength for Raman shifting to 589 nm. Initial results from YGAG indicate that it will not be a suitable material for creating the sodium guide star laser. Initial results from the YGAG laser is presented, along with a discussion of the potential of the technology.

Altair, the Gemini Adaptive Optics System, is currently being developed for use a the Cassegrain focus of the northern Gemini 8-meter Telescope, located on Mauna Kea, Hawaii. The Altair mechanical design must meet strict mass, center of gravity and volume constraints while providing a highly stable optical bench assembly, and housing and thermally controlling a large set of electronics. The optical bench design for Altair has been developed in tandem with the optical tolerance analysis to ensure that the structure could support the most sensitive optics within specification over the full range of gravity vector and temperature encountered. In turn,the optical tolerance budget was reallocated to better match the predicted mechanical performance of each optical element in the system. Although the system will initially be implemented with a natural guide star wavefront sensor, an upgrade path for a laser guide star wavefront sensor is being incorporated in the system design.

Over the past several years, a number of steps have been taken to improve the optical performance of the Vacuum Tower Solar Telescope at Sacramento Peak. We believe that the optical system of the telescope is now corrected to better than 1/10 wave rms after these improvements.

A tip-tilt control system has been built as part of the adaptive optics system for the Keck II telescope on Mauna Kea in Hawaii. This system is used to correct for wavefront tip-tilt when the adaptive optics system is in the laser guide star mode, and it uses a natural star as the reference. The system consists of a tip-tilt sensor, fast steering mirror, and digital controller. The tip-tilt sensor is based on a quadrant lens assembly with fiber-optics coupling to four photon counting silicon avalanche photodiodes. The fast steering mirror mount has three PZT actuators with position sensor, and an 8 inch Silicon Carbide lightweight mirror. The controller accommodates a range of integration times, and includes automatic light level control, and an adaptive control algorithm which optimizes control performance with changing tilt star image sizes. The design and performance characteristics of a tip- tilt control system for the Keck II telescope are presented.

Precise registration of telescope pupil, wavefront sensor camera, and science camera is critical to optimum performance of an adaptive optics system. For this reason great care is taken in design and construction to ensure optical rigidity and stability. However, some residual flexure is unavoidable, particularly for a system like the one being built at the JPL for the Cassegrain focus of the Palomar 200 inch, which will experience changing gravity loads as the telescope points and guides. Such flexure has direct impact when it occurs in the non-common-path segments of the wavefront-sensor and science beams. the diffraction- limited image size in K-band at the 5 m telescope is 30 microns. To maintain non-common-path flexure errors at a small fraction of this scale would require costly, gravity- compensated optomechanics. A more practical approach is to use precise measurements of the flexure in a periodic recalibration of the wavefront sensor offsets. As an alternative to mapping flexure over a grid of sky pointing positions, we have investigated a low-bandwidth metrology scheme in which we sense tilts and translations of optical components in the various light paths using visible laser signals. We find that such a system is relatively easy to implement, and can be expected to provide dramatic improvement in adaptive optics performance when the 'real- world' effects of telescope orientation are considered.

The WIYN 3.5 meter telescope is situated on the southwest ridge of Kitt Peak yielding excellent atmosphere seeing conditions. As such, the telescope and enclosure design was directed towards exploiting this feature. The primary mirror was spun cast and figured by the Steward Observatory Mirror Laboratory and the secondary mirror by Contraves. In both cases the performance exceeded the design specifications. The borosilicate primary is actively temperature controlled to within 0.2 C of the desired temperature, typically 0.5 degrees C below the ambient air. The telescope structure is also temperature controlled and the enclosure is opened to the outside ion all sides, which all heat sources are vented to ducts carrying air downwind of the facility. The primary mirror is actively controlled for low order aberrations by 66 axial actuators which are adjusted open loop via force matrix look-up tables and closed loop via real-time wavefront curvature sensing measurements. The active optics also included real-time collimation and focus control. The telescope drive and guider are capable of providing tracking to a few hundredths of a second of arc. By employing active telescope control at this level, it is possible to maintain telescope and local wavefront distortion to a level where atmospheric effects dominate the image quality. Since a significant fraction of the power in the atmospheric disturbances is contained in image motion the first step in adaptive optics control will be simple tip tilt. Studies of higher order AO system are being carried out, as well as additional test characterizing the telescope and site. It is intended to continue such studies in an attempt to establish long term variances.

We have developed and tested a method for minimizing static aberrations in adaptive optics systems. In order to correct the static phase aberrations, we need to measure the aberrations through the entire system. We have experimental setup demonstrating that phase retrieval can improve the static aberrations to below the 20 nm rms level, with the limiting factor being local turbulence in the AO system. Experimentally thus far, we have improved the static aberrations down to the 50 nm level, with the limiting factor being the ability to adjust the deformable mirror. This should be improved with better control algorithms now being implemented.

The influence of the turbulence outer scale on the power spectrum (PS) of speckle interferograms and on the Strehl ratio obtained with low-order adaptive optics (AO) system is examined by numerical simulation. It is shown that the outer scale has a little positive effect the PS when the outer scale magnitude becomes less than the aperture diameter. The finite outer scale has a positive effect on the STrehl ratio of the uncorrected long-exposure image. It is also shown that the Strehl ratio produced by a second-order AO correction is affected quite strongly by the finite outer scale. For the more higher-order correction, the effect under study is very weak and it appears only when the outer scale magnitude becomes less than the aperture diameter.

A Matlab front end to the LACE Adaptive Optics Performance Analysis program has been developed. The software can be run on any system that has the Matlab software installed. The graphical user interface for LACE is used to set the input parameters, to initiate execution, to monitor error messages during execution and to graphically illustrate the outputs of an individual runs as well as results over time.

At the Starfire Optical Range many different Adaptive Optics systems are configured and tried out, with many pieces of equipment supplied, programmed and supported by different people. To accommodate this dispersion of responsibility, while allowing centralized configuration, and operation, and ease of use and programming, a new software technology based on AppleEvents is utilized.

A 61-element adaptive optical system is being built for the 1.2m telescope of Yannan Observatory for astronomical observation in visible range. It consists of a 61-element deformable mirror, a fast steering mirror, a Hartmann-Shack wavefront sensor with 8 by 8 subapertures, a parallel digital processor and control electronics. The arrangement of subapertures is square matched with triable arrangement of actuators. The detector of H-S sensor is intensified CCD with frame rate of 838 Hz. The effective bandwidth is about 50 Hz for the DM control loop. A quadrant tracking detector consisting of 4 photon counting PMT detectors is used in the control loop of fast steering mirror. This system has been tested. The basic configuration and the preliminary results of the test are reported in this paper. This system will be integrated with a laser guide star system, in which a frequency doubling Nd:YAG pumped by laser diode array is used for producing a Rayleigh guide star. The pulse energy is 100 mJ with frequency of 838 Hz. A separated optics will be used to project the laser beam. The estimated specifications of this Rayleigh guide star system will be described in this paper too.

SINFONI, the SINgle Faint Object Near-IR Investigation, is an instrument for the very large telescope, designed to provide spectroscopy at the telescope diffraction limit in the near-IR. This unique capability is achieved by combining two state-of-the-art developments, an integral field spectrometer and a curvature sensor based adaptive optics system. SINFONI is a collaborative effort by the Max-Planck- Institut fuer extraterrsetrische Physik and the European Southern Observatory.

The University of Mexico 6.5m TIM telescope project has a 19 fully active segmented M1 and an active M2. Each 1.8m segment is a relative thin meniscus supported by three position hard points, computer controlled, and 16 force soft actuators which support the meniscus and can deform it. This project includes an AO system using a corrector mirror placed at a conjugated telescope pupil. We decided to specify the telescope optics in function of this AO system from the beginning and not specify the AO system in function of the telescope. In order to calculate the optical budget of the TIM we simulated, using a Monte Carlo technique e approach, the Strehl ratio of the telescope PSF as a function of the low order aberration values of the individual segments. The low order aberrations studied are the 'position' aberrations, as the piston and the tip-tilt between the segments, and the 'deformation' aberrations as the values for the deformation aberrations for each segment are comparable with the 'intrinsic' quality of the primary mirror of an active telescope as the NTT or our recently enhanced 2.1m UNAM telescope, if diffraction limited imaging for the complete primary is specified. This result shows that the technology for a correct support for the individual segments actually exist. The results obtained are used for the optical quality definition of the TIM 6.5m telescope.

We have been studying a high-bit-rate satellite laser communications system between a ground station and a geostationary satellite, named high-speed optical feeder- link system. An adaptive optics (AO) is the best solution to compensate the degradation of the laser beam quality due to atmospheric turbulence and to realize such a high-bit-rate laser communication system.

The Multiple Mirror Telescope on Mt. Hopkins will soon be upgraded to a single 6.5 m primary mirror.An adaptive- optical system, featuring a thin-shell, adaptive-secondary mirror with 330 voice-coil actuators is being developed for this new telescope. The thin-shell mirror is supported by a thick, concave aluminum substrate which also holds the actuator control and monitoring electronics and serves as the reference surface. With the actuator electronics dissipating heat into the substrate, the thermal behavior of the aluminum reference plate becomes as important issue. This paper presents results form tow experiments designed to determine the thermal behavior of the reference plate.

The design of an adaptive secondary mirror has to satisfy many requirements coming from different fields. The thin mirror must be actuated very precisely with a large bandwidth. The reference plate has to provide a high stability reference for the optical surfaces. The local seeing is not to be degraded by any significant thermal perturbation. In this article, the performances of a configuration with a single aluminum reference plate, that also provides the heat sink, are computed starting from the input power coming from the magnetic actuators, whose magnetic design has been revised.

An alternative approach to more compact deformable mirrors for adaptive optics is developed. A thin and flexible reflective membrane is coated with a magnetic layer and locally deformed by the field created by an array of planar microcoils. This novel technology should enable higher resolution with smaller, lighter integrated mirrors. Several complementary modeling tools are used to study the electromagnetic, mechanical, magneto-mechanical and thermal behavior of the device. A first prototype of (phi) 30 mm mirror with a matrix of permanent magnets and an array of 25 microcoils is under construction.

We designed and manufacture dan 'Adaptive Secondary Mirror 30 Actuators Prototype' that represents as much as possible the final MMT adaptive optics unit for what concerns mechanical, electronic and cooling system issues. Moreover, different actuator designs have been tested by using the same test bench. The adaptive secondary is based on a thin continuous facesheet mirror moved by electromagnetic actuators. Permanent magnets are glued on the mirror, while voice coils are connected to the support structure by cold fingers. A capacitive sensor co-located with each actuator measures the local gap between the mirror and a reference back plate. Voice coil motors allow to perform chopping and tip-tilt correction by using directly the deformable mirror. Actuators pitch is set to achieve high order correction in the visible wavelength.

ELECTRA is the University of Durham's high order AO system designed for the William Herschel Telescope. The correcting element is a segmented deformable mirror, where each segment is driven by a cylindrical piezoelectric actuator such that it can be moved in tip, tilt and piston. In order to cophase all the segments of the mirror it is necessary to calibrate and linearize the response of these actuators. To this end, we have installed strain gauge transducers along the electrodes of each actuator to provide information on their instantaneous extensions and calibrated their responses.

The wavefront correction by an adaptive optics (AO) system is never perfect, but images obtained with AO can be restored by deconvolution methods, if the point spread function is known. We present the first result of PSF reconstruction from the wavefront sensor measurements applied to the AO system ADONIS installed at the 3.6m ESO telescope at La Silla.

A Lorentzian shaped point spread function (PSF) for adaptive optics is demonstrated. In some cases, a Lorentz atop an Airy pattern is a better description of the PSF. If the object can also be described with an analytic function, then Parametric Blind Deconvolution can be applied as a self referencing method to extract information about both the object and PSF simultaneously. The technique is used to derive the triaxial shape and rotational pole of the asteroid Vesta from observations made on the night of October 18, 1997, marking the first time the high order loop was closed for science with the Starfire Optical Range 's 3.5m telescope.

In this paper, we present a novel approach to modeling Shack-Hartmann-based adaptive optics (AO) system which allow to easily predict their performance. The idea is to start for the power spectral density (PSD) of the turbulent phase and to derive the PSD of the compensated phase by taking into account the various errors affecting the AO correction, which can be expressed analytically in the Fourier domain. Once the PSD of the compensated phase is known, the computation of the residual point spread function is straightforward.

While still in their infancy, adaptive optics systems give rich promise of providing to ground-based and space-based observers the capability to obtain near-diffraction limited images through the earth's atmosphere or though new lower- cost space optical systems. Close study of the images now being obtained with AO systems on ground-based systems, aiming toward the serious photometric use of the current systems, shows that serious image artifacts exist. These affect both the study of morphology and the photometric accuracy. This paper explores the source of these artifacts by analytic modeling of the AO systems, and compares the result of the modeling with data obtained on the telescopes.

We describe the current planning and activities for the laser Guide Star Facility of the VLT observatory. The current activity aims at a feasibility study and to the experimental verification of innovative ideas in the choice of the laser and in the relay system.

We show how to resolve the autocorrelation of wavefront- sensor measurements as they are commonly taken by adaptive- optics systems into terms due to individual atmospheric layers. We exploit this factorization to measure the strength and wind speed of each turbulent layer.We show that the Kolmogorov theory provides a good description of the data and find values of Fried's parameter, r₀, that are in good agreement with other measurements. Finally, a quantitative estimate of the validity of Taylor's frozen flow hypothesis is obtained.

In this paper we compare the impact on the turbulent wavefront Zernike modes, of a Kolmogorov and a Von Karman spectrum atmosphere with finite outer scale. We have simulated these effects analytically and numerically, for Natural and Laser Guide Star AO system, on axis and off- axis. The goal is to compare the effects of anisoplanatism, and re-evaluate the discussions on the pro's and con's of LGS vs. NGS on a perspective which includes also the presence of the outer scale.

We have obtained high speed image motion data from the 3.5M WIYN telescope at Kitt Peak as part of commissioning and characterizing efforts. These data come from a small frame transfer CCD feeding dedicated centroiding hardware called FastTrack, which provides x and y data pairs at frequencies > Hertz. In this paper we use power spectra from these data to investigate telescope structure resonance and characterize the site + telescope performance for gains by implementing tip-tilt correction at WIYN. At frequencies below 20hz the image motion power spectra are consistent with Kolmogorov turbulence models but show excess power at higher frequencies. Over the past 2 years the FastTrack data have consistently shown that image improvements of 0.10-0.20 arcseconds can be obtained with a simple tip-tilt systems that fully corrects frequencies 20Hz and lower. The FastTrack power spectra have also revealed a complex structure of coherent frequencies between 22 and 28 Hertz, similar to what has been seen on other light weight stiff telescopes. In other tests, we have used simultaneous star trail data to estimate the isokinetic angle for frequencies below 10hz. We have found that the one dimensional correlation remains above 0.8 within an angular radius of approximately 240 arcseconds. These data show that a high speed tip-tilt system can net significant improvements to image quality at WIYN over a relatively large field.

We have pointed out that such two terms as 'an effective scattering volume' and 'a laser guide star' (LGS) are scientific synonyms. The first term had been introduced earlier by specialists in atmospheric optics and laser sounding. In the report we have also represented information concerning fluctuations of waves reflected from an object and waves passed twice atmospheric inhomogeneities. Corresponding mathematical apparatus allows one to estimate correlation characteristics of a LGS. We have considered the mutual correlation function of random angular displacements of a plane wave image and the centroid displacements of a focused Gaussian laser beam. The algorithm of 'optimal' correction was introduced in which a priory information is employed. Basing on this algorithm the variance was evaluated of residual jitters of a star image obtained with the use of LGS.

We discuss here some possibilities of introducing the distortions of an optical wave phase, propagating along vertical atmospheric path, as an integral characteristic, describing the turbulence along the path. Several models of the turbulence outer scale profile have been analyzed as well as the structural characteristic of the atmospheric refractive index fluctuations in order to find the value of the efficient outer scale. The error in the Strehl ratio determination was estimated. This ratio was computed using the efficient outer scale and compared to its value computed by the model profile of the outer scale.

At the INSTITUTO DE ASTRONOMIA we developed an adaptive optics system for the correction of the two first orders of the Zernike polynomials measuring the image controid. Here we discus the two system modalities based in two different control strategies and we present simulations comparing the systems. For the classic control system we present telescope results.

High resolution adaptive phase distortion suppression was experimentally demonstrated for laser interferometers using a liquid crystal television as a phase modulator and an opto-electronic feedback loop. The experiments were carried out for both amplitude division and rotational shear types of interferometers. The suggested iterative control algorithm is based solely on interference pattern information. The results of numerical simulations of a high resolution adaptive system based on rotational shear type interferometers show the system's potential for atmospheric turbulence phase distortion suppression.

There are considerable potential advantages in throughput, polarization and IR emissivity of performing the adaptive correction at the telescope secondary. However, reliability and safety are key issues. This paper explores a sound engineering approach which is rugged, safe and provides a level of redundancy. A 7-actuator demonstrator is described, together with first results from laboratory testing.

GIven are the results of experimental study on the quasi real time holographic correction for the lens distortions in the passive observational telescope in the visible range of spectrum, using the liquid crystal optically addressed spatial light modulator.

Large numerical aperture telescope with nonlinear optical correction for distortions, designed for the remote self- luminous object imaging, was realized in experiment and investigate. Dynamic hologram, recorded in optically addressed liquid crystal spatial light modulator, was used as the corrector. Nearly diffraction limited performance of the system was demonstrated.

AdOptTNG is the Adaptive Optics Module for the Galileo National telescope (TNG). At first light this module will feature tip-tilt correction capabilities and a speckle interferometry facility. In this paper we describe the main characteristics of the AdOptTNG Control System Software. The system is composed by a real-time part and a graphical user interface. The real-time software computes the required corrections, drives the tip-tilt mirror and function. The user interface has been written in IDL, using the IDL Widget set, taking advantage of its powerful data manipulation and analysis capabilities for data presentation and diagnostics. The integration of all these components into the TNG Control System Software, in order to allow for remote control and data archiving, is also briefly described.

Many factors contribute to the aberrations induced in an optical system. Atmospheric turbulence between the object and the imaging system, physical or thermal perturbations in optical elements degraded the system's point spread function, and misaligned optics are the primary sources of aberrations that affect image quality. The design of a non- conventional real-time adaptive optic system using a micro- mirror device for wavefront correction is presented. The adaptive optic system uses a VLSI circuit that can be reconfigured for use with many wavefront sensor including the Hartmann, shearing, and curvature wavefront sensors. The unconventional adaptive optic imaging systems presented offer advantages in speed, cost, power consumption, and weight. Experimental and modeling results that characterizes the performance of each wavefront sensor in the micro-mirror adaptive optic system are presented.

Observations using the Starfire Optical Range (SOR) 1.5m telescope, located near Albuquerque, NM, were made during two separate observing runs, one in 1995 and the other in 1996. Image data was collected using a cooled 2K by 2K user provided CCD camera system. During the first observing sessions a standard SOR direct imaging configuration was used where a CCD imager. For the second observing run the configuration was modified to use a mirror with a small on- axis hole that allowed all the light in the central region to be used by the AO system while the remainder of the field was directed to the CCD. The data from these observations were used to investigate a number of issues related to AO observing including: (i) the effect of scattered laser light on image quality; (ii) the photometric accuracy across an AO corrected field; (iii) the PSF variations across an AO corrected field; (iv) the limits to observation of close companions using a mirror with a hole as a coronograph. The result of these observing runs are presented along with representative images obtained using no correction, partial correction, using natural guide stars, and Rayleigh laser beacon configuration.

A new silicon deformable mirror design is presented which provides high reflectivity, an optical quality continuous surface, and high intensity handling capacity. Its features make it useful for a wide range of devices including telescopes, leasers, and photolithographic systems. The mirror architecture is similar to commercial electrostrictive deformable mirrors. A focus corrector built using this architecture exhibited 2.25 microns of actuation at the center through the application of 100v corresponding to a radius of curvature of -2.4m. A 30 micrometers thick 1cm diameter silicon mirrors exhibited its first mechanical resonance at 2.7kHz. A mirror coated with 100nm of gold was shown to be able to withstand 100kW/cm² of continuous wave 1064nm intensity for 10 minutes without observable degradation. An active mode-matching experiment was performed showing that 99.5 percent of a Nd:YAG beam could be coupled to a finesse 4000 ring cavity.

We report on the performance of silicon micromachined adaptive mirrors in a closed-loop system designed for correction of primary optical aberrations. The feedback was configured to optimize for the beam brightness, which resulted in efficient minimization of on-axis aberrations. To characterize the performance we used independent interferometric and far-field monitoring of the beam quality. The optimization system provided suitable for high quality quasi-static correction of low-order aberrations with amplitudes of up to several wavelengths. The approach reported has a practical potential in applications ranged from retinal medical imaging to on-board space-based laser communication systems and telescopes.

We present a rapid technique for the simultaneous calibration and flattening of a deformable mirror in an adaptive optics system. We achieve an rms surface figure equivalent to better than (lambda) /100 in the K band in our first application of the scheme. The technique requires the presented of a phase shifting interferometer within the adaptive optics system. Measurements of the mirror surface are made while each mirror actuator's gain is estimated, in a stable, system. Measurements of the mirror surface are made while each mirror actuator's gain is estimated, in a stable, iterative, convergent sequence. Knowledge of the mirror responses function to single actuator activation is not needed. The iterative process ends when the mirror surface reaches some preset desired flatness, which can be specified by the rms deviation of the surface. An overall scale correction to the gains is measured and applied after the iterations converge. Our technique uses the mirror under conditions similar to those present when the adaptive optics system is running. This technique provides the user the ability to calibrate the mirror in a matter of a minute or two, enabling re-calibration of the actuator gains may times during a night with minimal impact on observing duty cycle, in order to adapt to changing conditions.

We report on first observation run with the Achromatic Interfero Coronagraph (AIC) developed at Observatoire de la Cote d'Azure, France. Observations took place last Fall at Observatoire de Haute Provence with the 1.52 m telescope equipped at that time with adaptive optics. The AIC is an imaging device providing the nulling of a star without nulling the close environment of this star. Nulling results from a destructive interference process. Morphological features located as close to the star as the first angular Airy ring can be detected, thus breaking a limitation of the classical Lyot coronagraphs. The objectives of the observation run is to demonstrate that the AIC can image faint companions very close to the diffraction limit with ground-based telescope. After a short reminding of the principle of the AIC, conditions of observations are reported and first coronagraphed-images are shown. Finally limitations are discussed and improvements to carry on are described.

The effective wavefront outer scale L₀ was monitored continuously during 16 nights in August-September 1997 using the new Grating Scale Monitor (GSM) instrument. Simultaneous comparison with the data form open-loop adaptive optics system on the ESO 3.6 m telescope revealed a good agreement of the L₀ derived by both methods. The GSM principle is similar to the Shack-Hartmann sensor: angle of arrival fluctuations are measured with four 10-cm telescopes and the outer scale is computed from their normalized covariances. The seeing and effective wavefront velocity are measured as well. GSM does not suffer form dome and mirror seeing effects and can be used anywhere for site evaluation. Outer scale has log-normal distribution, its median is found to be 24 m, with a rms scatter of log L₀ of +/- 0.22. Short isolated bursts of high L₀ values were sometimes observed. A weak correlation of L₀ and seeing angle exists for some nights. Decametric values of L₀ found here imply that at baselines of few meters the influence of finite L₀ on the energy balance between tip-tilt and higher-order modes must be properly taken into account. This can sensibly modify the predicted performance of adaptive optics systems.

Adaptive Optics Associates (AOA) has recently developed a commercial Shack-Hartmann wavefront senor called WaveScope for use in optical testing and adaptive optics. The sensor head and associated wavefront analysis software are a powerful and highly flexible combination. Wavefront data can be manipulated and displayed using Tk/Tcl commands and AOA's own atomic functions. To demonstrate this over the last month we have integrated a deformable mirror manufactured by OKOTechnologies into WaveScope. All the software necessary to control the mirror and perform closed loop adaptive optics was written in the Tk/Tcl scripting language which ships with WaveScope. The result is a low cost integrated adaptive optics system.

We present diffraction-limited near-IR images in J, H and K of the nucleus of several AGNs obtained with PUEO, the Adaptive Optics system of CFHT. Thanks to the excellent sensitivity of the wavefront sensor, galactic nuclei as faint as mV equals 15 can be observed at a resolution of 0.12- 0.18 inch. IN NGC 1068, the Seyfert 2 prototype, the achieved resolution reveals: a) a belt which is most probably the trace of warm dust in the torus of gas and b) a S-shaped feature which is likely an extremely compact barred spiral structure. In Mkn 231, an Ultra Luminous galaxy, we resolve the very central peak in a bright point source with colors of an actual AGN and fainter source (B) at 0.15 arcsec north, with colors consistent with a giant HII complex of violent star formation. Arp 299, another Ultra- Luminous IR Galaxy, is a merger. Our K image reveals that none of the nuclei are un resolved; the presence of a QSO as the primary source of energy is thus dubious. We also discovered many super-clusters of newly formed stars with extremely high luminosities.