Over the past two years ESO has reinforced its efforts in the field of Adaptive Optics. The AO team has currently the challenging objectives to provide 8 Adaptive Optics systems for the VLT in the coming years and has now a world-leading role in that field. This paper will review all AO projects and plans. We will present an overview of the Nasmyth Adaptive Optics System (NAOS) with its infrared imager CONICA installed successfully at the VLT last year. Sodium Laser Guide Star plans will be introduced. The status of the 4 curvature AO systems (MACAO) developed for the VLT interferometer will be discussed. The status of the SINFONI AO module developed to feed the infrared integral field spectrograph (SPIFFI) will be presented. A short description of the Multi-conjugate Adaptive optics Demonstrator MAD and its instrumentation will be introduced. Finally, we will present the plans for the VLT second-generation AO systems and the researches performed in the frame of OWL.

The purpose of this paper is to report on new adaptive optics (AO) developments at the W. M. Keck Observatory since the 2000 SPIE meeting. These developments include completion of the Keck I AO system, interferometric combination of the full apertures of the two Keck telescopes using AO on both telescopes, commissioning of two science instruments with the Keck II AO system, first projection of the Keck II sodium laser beacon, progress on laser guide star AO, improved automation of the AO systems and a diversity of AO science programs.

We present an overview of Subaru Cassegrain adaptive optics system and its performance verified at the engineering run. The system is based on a curvature wavefront sensor with 36-element sub-apertures and a bimorph deformable mirror with identical number of elements. We had the first light in Dec. 2000. The AO system has been in service for two instruments, IRCS; infrared camera and spectrograph, and CIAO; coronagraph imager with adaptive optics. The Strehl ratio at the K band is around 0.30 under 0.4- 0.5 arcsec K-band seeing condition for bright guide stars. The sensitivity of the wavefront sensor is so high that we have significant improvement of image quality even for a faint guide star down to R=18^th magnitude. The measurement of stars in a globular cluster suggests an isoplanatic angle, about 40 arcsec, wider than that expected from the equivalent turbulence layer assumed at the height of 6.5 km. The system has been offered for common use since Apr. 2002. Some scientific results using this AO system are shown in this paper.

A 36-elements curvature adaptive optics (AO) system has been operating on the Subaru telescope for about one and a half year. We achieved a Strehl ratio of 0.3 in the K-band, which is a rather smaller value than we expected. While we are investigating the discrepancy between the obtained performance and the simulated performance of the current AO system and we are also improving the current AO system in terms of the Strehl performance and the observing efficiency. Meanwhile we have started to plan a next generation of Subaru AO system. Two major upgrades are proposed in this paper. One is to increase the number of subapertures as much as possible. Practically, the number of subapertures lies between 100 and 200. The size of subaperture becomes half to one-third of that of the current system and we expected that the K-band Strehl ratio will improve to more than 0.6. The first light of the higher order curvature AO system is scheduled for 2004. Another upgrade plan is to use a laser guide star (LGS). A single LGS is projected at the sodium layer with an output power of 4 W. Conceptual designs for the laser system, beam relay system, laser launching telescope and control system have begun. The first test of launching laser from Subaru telescope will be in 2005.

A laser guided adaptive optics system called UnISIS -- University of Illinois Seeing Improvement System -- has been commissioned at the Mt. Wilson 2.5-m Telescope. The UnISIS laser guide star is created via Rayleigh scattering of 351 nm photons from a 30 W excimer laser. The laser guide star is focused at an altitude of 20 km above msl. The UnISIS adaptive optics system sits at the fixed f/30 Coude focus of the 2.5-m telescope while the 30 W excimer laser sits on the observatory ground floor. The collimated laser beam is projected first into the Coude room where it is converted to f/30 and projected into the sky off the 2.5-m primary mirror. We describe the practical experience gained in installing and commissioning UnISIS, and we present simulations of the expected system performance based on the characteristic C_n² distribution above Mt. Wilson.

The multi-conjugate adaptive optics (MCAO) system design for the Gemini-South 8-meter telescope will provide near-diffraction-limited, highly uniform atmospheric turbulence compensation at near-infrared wavelengths over a 2 arc minute diameter field-of-view. The design includes three deformable mirrors optically conjugate to ranges of 0, 4.5, and 9.0 kilometers with 349, 468, and 208 actuators, five 10-Watt-class sodium laser guide stars (LGSs) projected from a laser launch telescope located behind the Gemini secondary mirror, five Shack-Hartmann LGS wavefront sensors of order 16 by 16, and three tip/tilt natural guide star (NGS) wavefront sensors to measure tip/tilt and tilt anisoplanatism wavefront errors. The WFS sampling rate is 800 Hz. This paper provides a brief overview of sample science applications and performance estimates for the Gemini South MCAO system, together with a summary of the performance requirements and/or design status of the principal subsystems. These include the adaptive optics module (AOM), the laser system (LS), the beam transfer optics (BTO) and laser launch telescope (LLT), the real time control (RTC) system, and the aircraft safety system (SALSA).

The adaptive optics system for the Gemini South telescope, currently in the design phase, consists of several major subsystem. The largest subsystem, called the AO module, contains most of the optics and electronics and is mounted on one of the Cassegrain instrument ports. The initial system will be a conventional laser guide star AO system, but the plan is to eventually expand it to a multi-conjugate system. The system is being designed to readily add the components necessary to upgrade to a multi-conjugate system. This paper describes the design challenges encountered and solutions that were derived for the AO module design. The complexity of the multi-conjugate version is illustrated, including optical, mechanical, electronic and controls issues.

Since a few years we are developing a low cost adaptive optics system for the GI2T optical interferometer. Our AO is based on a curvature sensor and a 31 actuator bimorphe mirror. We designed a wavefront sensor using an array of prisms to split the pupil image and photon counting avalanche photodiodes modules as detectors. We present here the design and the first results obtained on a test bench. This AO system will be used for the tests of a laser guide star and for observations of stars and asteroids. Latter, another system will be built to equip GI2T-Regain with two AO.

The goal of this paper is to discuss the turbulent effects on the laser beam of the lunar laser ranging (LLR), and to use the real-time tilt compensation for the LLR. In this new technique, a small area near the retroreflector array on the moon surface will be used as an extended source to detect the wavefront tilt. Then the absolute difference algorithm will be used to calculate the wavefront tilt and to reconstruct it, and it will be used to drive a tip-tilt mirror in the laser transmit path to carry on the real-time tilt compensation for the laser beam on the LLR. After the compensation, the returned laser photons from the moon retroreflector to the ground telescope will be increased a factor of 6 to 40, depend on the turbulence.

We discuss the image improvement from an adaptive optics system that selectively corrects for phase perturbations with large angular coherence lengths. Selecting phase perturbations with large angular coherence lengths effectively sets the correction to turbulence of all spatial frequencies at the ground and to turbulence of progressively lower spatial frequencies at higher altitudes. The performance of a high-order curvature system (85 elements) on a small telescope (4 meters) across a field of view of roughly six arcminutes diameter is simulated using Monte Carlo simulations. The implementation studied assumes a single correcting element conjugate to the ground with multiple wave front sensors each pointed at a different star within the field of view. The image full-width at half maximum (FWHM) is reduced by up to a factor of two over the uncorrected case. Even though the number of guide stars is large (n~3), the sky coverage is similar to that for a classical adaptive optics system since the acquisition field is much larger (~35 square arcminutes). In principle the approach could be extended to multiple laser guide stars and larger apertures. The requirements for the laser are relaxed since the sampling of turbulence is preferentially at low-altitudes (e.g. Rayleigh beacons).

Affordable adaptive optics on small telescopes allow to introduce the technology to a large community and provide opportunities to train new specialists in the field. We have developed a low order, low cost adaptive optics system for the 1.6m telescope of the Mont Megantic Observatory. The system corrects tip-tilt, focus, astigmatisms and one trefoil term. It explores a number of new approaches. The sensor receives a single out-of-focus image of the reference star. The central obstruction of the telescope can free the focus detection from the effect of seeing and allows a very small defocus. The deformable mirror is profiled so as to preserve a parabolic shape under pressure from actuators located at its edge. A separate piezoelectric platform drives the tilt mirror.

We have designed, realized and tested a dedicated software tool defined so as to enable a wide non-specialist community to perform optimal adaptive optics observations with VLT instrument NAOS-CONICA. We first precise the requirements derived from NAOS complexity due to a large number of configurations, ESO/VLT operational policy including service mode observations and limited human interaction during observations, and astronomical observation requirements. We then present the developed software tool, so-called "Preparation Software", that couples a user-friendly interface that accepts observation conditions (including seeing, star magnitude etc...) and an elaborate simulator of adaptive optics based on the NAOS characteristics.

Use of a highly sensitive curvature wavefront sensor (WFS) with avalanche photodiode (APD) detectors at the Gemini North telescope has allowed direct AO guiding on very low mass stars (V=18-20, I=14-16) producing 0.1” images in K’. This resolution (which is near to the 0.07” diffraction limit) has enabled us to conduct ground-based searches for substellar companions (<0.075 M) to within ~3 AU of their primary stars. We have been able to use the faint science targets themselves to close the AO loop. The key to this capability is the zero readout noise, photon-counting APDs used in the curvature WFS. The amplification of incident photons from low luminosity sources allows sufficient signal at the subapertures to reconstruct the aberrated wavefronts with relatively little reconstruction error. This is advantageous compared to typical Shack-Hartmann WFSs which require greater signal from their guide stars to compensate for their CCDs’ readout noise. The technique presented here is currently the only working one from the ground studying such cool and faint targets. Additionally, we report the discovery of 3 new binary stellar systems from a survey of 30 low mass stars (~0.095 M, spectral type M6.0-M7.5, V=18-20) at separations between 0.12-0.29” (3.5-8.3 AU) using this technique.

At TNO TPD we have realized an Adaptive Optics test bench. The bench has an in-house built turbulent atmosphere simulator. For the wavefront sensors there is the possibility to choose between a Shack-Hartmann sensor and a pyramid sensor. Compensation of the wavefront error is performed by a separate tip-tilt mirror and a deformable mirror. Both are off the shelf products. The AO system is controlled by a Multi-Input-Multi-Output control system with 40 actuator channels and 50 sensor channels. The proposed control strategy corresponds to a Linear Quadratic approach, in which the sum of the mean squared wavefront error over the sensor points and the weighted control effort is minimized. In the optimization process the dynamic properties (spatial-temporal correlation) of both the turbulence-induced wavefront error and the mirror/sensor combination are taken into account. Furthermore, important issues like closed-loop stability and robustness are included in the control design. An adaptive control algorithm has been derived, which converges to the LQ-solution and also enables tracking of changes in the characteristics of the turbulent wavefront. This paper presents the first results achieved with the Adaptive optics test bench. It shows that a simplified version of the adaptive feedback control strategy already gives promising results, both implemented in a commonly used AO simulation software package and in real-time on the AO test bench.

NAOS is the first adaptive optics system installed at the VLT 8m telescopes. It was designed, manufactured and tested by a french Consortium under an ESO contract, to provide compensated images to the high angular resolution IR spectro-imaging camera (CONICA) in the 1 to 5 μm spectral range. It is equipped with a 185 actuator deformable mirror, a tip/tilt mirror and two wavefront sensors, one in the visible and one in the near IR spectral range. It has been installed in November at the Nasmyth focus B of the VLT UT4. During the first light run in December 2001, NAOS has delivered a Strehl ratio of 50 under average seeing conditions for bright guide stars. The diffraction limit of the telescope has been achieved at 2.2 μm. The closed loop operation has been very robust under bad seeing conditions. It was also possible to obtain a substantial correction with mV=17.6 and mK=13.1 reference stars. The on-sky acceptance tests of NAOS-CONICA were completed in May 2002 and the instrument will be made available to the European astronomical community in October by ESO. This paper describes the system and present the on-sky performance in terms of Strehl ratio, seeing conditions and guide star magnitude.

Observing at high angular resolution from the ground is not made possible with Adaptive Optics alone, and besides the turbulence residuals, atmospheric refraction, thermal background or instrument's mechanical flexures may also severely limit the gain of optical quality that AO techniques are supposed to provide. We describe here how NAOS, the newly installed AO system on the VLT, has been designed to accommodate for these unavoidable effects. In particular, beam chopping, flexures compensation and AO tracking on reference objects with a significant relative motion will be addressed. It will thus be shown how long term astronomical observations at the diffraction limit can be carried out with an AO system under regular ground level conditions, thanks to the implementation of original technical solutions.

In this communication, we present the progress of the 6.5m MMT adaptive optics system. During the last part of 2001 and the 1^st part of 2002, the system has been validated in the laboratory statically and dynamically with sample frequencies of up to 550 Hz. In June 2002, an attempt has been made to make this system work on the telescope but has been hampered by mechanical failures. However, ease of installation of the system and open-loop operation of the mirror was demonstrated at this occasion and offers reasons to be optimistic on the future of the system. The MMT-AO system is the first AO system to compensate the aberrated wavefront at the telescope's secondary mirror. This approach has unique advantages in terms of optical simplicity, high throughput and low emissivity. Its realization presents many technical challenges, which have now been overcome. Today, the deformable mirror is characterized and accepted. It features a 1.8 mm thick 640mm diameter convex aspheric mirror (manufactured at the Steward Observatory Mirror Lab), mounted on a 50 mm thick ULE reference body with 336 actuators, as well as a cluster of 168 DSP’s and associated analog circuitry for position sensing and actuator driving. The system has been characterized in the laboratory at sampling speeds up to 550 Hz and had been integrated on the telescope.

The paper describes the design of the single conjugate Adaptive Optics system to be installed on the LBT telescope. This system will be located in the Acquisition, Guiding and Wavefront sensor unit (AGW) mounted at the front bent Gregorian focus of LBT. Two innovative key features of this system are the Adaptive Secondary Mirror and the Pyramid Wavefront Sensor. The secondary provides 672 actuators wavefront correction available at the various foci of LBT. Due to the adaptive secondary mirror there is no need to optically conjugate the pupil on the deformable mirror. This allows having a very short sensor optical path made up using small dimension refractive optics. The overall AO system has a transmission of 70 % and fits in a rectangle of about 400×320mm. The pyramid sensor allows having different pupil sampling using on-chip binning of the detector. Main pupil samplings for the LBT system are 30×30, 15×15 and 10×10. Reference star acquisition is obtained moving the wavefront sensor unit in a field of view of 3×2 arcmin. Computer simulations of the overall system performance show the good correction achievable in J, H, and K. In particular, in our configuration, the limiting magnitude of pyramid sensor results more than one magnitude fainter with respect to Shack- Hartmann sensor. This feature directly translates in an increased sky coverage that is, in K band, about doubled with respect to the same AO system using a Shack-Hartmann sensor.

MACAO stands for Multi Application Curvature Adaptive Optics. A similar concept is applied to fulfill the need for wavefront correction for several VLT instruments. MACAO-VLTI is one of these built in 4 copies in order to equip the Coude focii of the ESO VLT's. The optical beams will then be corrected before interferometric recombination in the VLTI (Very Large Telescope Interferometer) laboratory. MACAO-VLTI uses a 60 elements bimorph mirror and curvature wavefront sensor. A custom made board processes the signals provided by the wavefront detectors, 60 Avalanche Photo-diodes, and transfer them to a commercial Power PC CPU board for Real Time Calculation. Mirrors Commands are sent to a High Voltage amplifier unit through an optical fiber link. The tip-tilt correction is done by a dedicated Tip-tilt mount holding the deformable mirror. The whole wavefront is located at the Coude focus. Software is developed in house and is ESO compatible. Expected performance is a Strehl ratio sligthly under 60% at 2.2 micron for bright reference sources (star V<10) and a limiting magnitude of 17.5 (Strehl ~0.1). The four systems will be installed in Paranal successively, the first one being planned for June 2003 and the last one for June 2004.

The design of a detector array intended for use as a Shack-Hartman wavefront sensor is presented. The chip will output voltages that represent local tilts and currents that represent inferred piston information by processing the optical signals from a lenslet array. Each pixel in the array contains four phototransistors, arranged in a square, along with transistors needed to perform algorithms necessary to output signals. Each pixel has four built-in amplifiers to output a voltage for each of the four phototransistors. By combining these voltages off-chip, tilt information is obtained, and allows for positive and negative tilts to be output from a single-supply chip. All pixels contain a self-biasing circuit, allowing tilt information to be consistent across all pixels for apertures that are not uniformly illuminated. Algorithms for inferring piston information are generated by using CMOS transistors in current mode. The piston information is output as a current with a magnitude that is proportional to the actual piston value. The pixels are selected by means of bit-parallel row and column select. The outputs consist of four voltages, four currents to monitor the phototransistor behavior, and one current representing piston. There are inputs that allow the user to set a global piston offset, and overall bias for overriding the amplification of the tilt voltages. Simulations of this design suggest refresh rates in excess of 2kHz are easily attainable. The design of this wavefront sensor is optimized for low light levels and high refresh rates. Prototype detectors will be fabricated using a 0.5micron CMOS process. This detector could greatly simplify the process of wavefront reconstruction, and could even allow for direct hardware control of deformable mirrors in adaptive optics systems with an order of magnitude more channels than are currently used.

Only a very few examples of near-infrared wavefront sensors can be found in the litterature. However, none of these sensors provide routine observation yet. Our sensor is the only one to be operated routinely on a large AO system. Entirely cryogenized, this sensor is built around a so-called HAWAII array from Rockwell (HgCdTe, 1024×1024). It is working in the huge spectral band ranging from 0.8 to 2.55 microns, and may use -when required- all the flux from this very whole band. It allows to switch between several optical configurations in order to match all atmospheric and observing conditions, while its original mechanical design allows to keep, even at cryogenic temperatures, a mechanical stability lower than 4 microns in any position. It also has some particular read-out schemes, allowing to obtain frame rates as high as 1200 Hz while keeping a read-out noise performance of 10 electrons rms/pixel. The analysis of the design parameters (pixel size, field of view) is exposed in this article. Some results, obtained during the comissioning runs at ESO, will also be presented.

In an adaptive optical system, it is essential that the wavefront sensor be accurately calibrated and aligned to the wavefront corrector. For the case of the Shack Hartmann sensor, there are at least three quantities which must be measured: a. the response of the subapertures to local wavefront tilt, b. the location of the zero point for each subaperture from a plane wave input, and c. the spatial relationship between the wavefront sensor subapertures and the wavefront corrector actuators. This paper present a method which verifies the calibration of a Shack Hartmann wavefront sensor by simultaneously measuring static wavefront with a phase shifting interferometer. These measurements were made using an apparatus we constructed in the laboratory to build and test the adaptive optical system for the 6.5 m MMT.

The estimation accuracy of wavefront sensors in strong scintillation is examined. Wave optical simulation is used to characterize the performance of several wavefront sensors in the absence of measurement noise. The estimation accuracy of a Schack-Hartmann sensor is shown to be poor in strong scintillation due primarily to the presence of branch points in the phase function. The estimation accuracy of a unit-shear, shearing interferometer is found to be significantly better than that of a Hartmann sensor in strong scintillation. The estimation accuracy of a phase shifting point diffraction interferometer is shown to be invariant with scintillation.

Use of ground-based observatories for astronomy has been greatly improved through the use of active correction systems based on subaperture wavefront sensing techniques such as Hartmann and curvature sensing. While the resulting performance significantly exceeds the performance of the unaided observatory, the use of a large number of independent subapertures to determine the necessary local corrections results in a requirement for relatively bright guidestars. This, in turn, limits the available area of the sky that supports high quality imagery of target using natural guidestar operation. The use of artificial guidestars increases the available region of the sky for viewing, but the resulting imagery is not as good as is achievable in the vicinity of a natural guidestar. The use of image-based sensing such as focus diverse phase retrieval produces higher quality wavefront sensing, but, if used as the sole WFS machine, generally requires longer processing times than can used for real-time correction. In this paper, we discuss the potential of using secondary wavefront sensing and correction systems within individual instruments to supplement the observatory active system. In particular, we present simulation results demonstrating the performance of a potential real-time focus diverse phase retrieval based WFS&C subsystem. We discuss the required observatory active correction performance, the secondary guidestar characteristics, and the processing speed requirements.

A new tracker is under development at the Starfire Optical Range (SOR). The tracker system's tilt sensor is based on an optical pyramid that creates a quad-cell effect into avalanche photo diodes, sampled at 5000 frames per second. A suite of hardware and software that allows unprecedented flexibility in control system design and implementation delivers the mirror control signals. The control problems associated with this tracking system are interesting due to lightly damped mirror resonances; these dynamics are handled via a tight inner cage loop. Thus, the controller makes use of both optical feedback and mirror position feedback. This paper discusses the design and implementation of the entire tracker system, including sensor characteristics, steering mirror characteristics, electronics, and control algorithms implemented. Experimental results are presented.

The NAOS adaptive optics system was installed in December 2001 on the Nasmyth focus of the ESO VLT. It includes two wavefront sensors: one is working at IR wavelength analysis and the other at visible wavelengths. This paper describes the NAOS Visible Wave Front Sensor based on a Shack-Hartman principle and its performances as measured on the sky. This wavefront sensor includes within a continuous flow liquid nitrogen cryostat: 1) a low noise fast readout CCD camera controlled by the ESO new generation CCD system FIERA using a fast frame rate EEV/Marconi CCD-50 focal plane array. This 128×128 pixels focal plane array has a readout noise of 3 e- at 50 kilopixel/sec/port. FIERA provides remotely controlled readout modes with optional binning, windowing and flexible integration time. 2) two remotely exchangeable micro-lens arrays (14×14 and 7×7 micro-lenses) cooled at the CCD temperature ( -100 °C) within the cryostat. The CCD array is directly located in the micro lenses focal plane at a few millimeters apart without relay optics. 3) Additional opto-mechanical functions are also provided (atmospheric dispersion compensator, flux level control, field of view limitation). On sky performances of the wavefront sensor are presented. Adaptive Optics corrections was obtained with a reference star as faint as a visible magnitude 17 with a band-path of 40 Hz in close loop.

In Astronomy the goal of the Adaptive Optics systems is the real time correction of the aberrations introduced from the turbolence of the air in the wavefront of the observed field of view. A wavefront sensor with pyramidic shape has being developed by a group of Italian researchers that offers the advantage of either variable gain against the wavefront deformation and tunable sampling of the telescope pupil. Single pyramid prototypes were made using the classical figuring and polishing techniques. This approach however, is not only very time consuming but also does not guarantee a uniform repeatability of the optical characteristics of the pyramids requested by this application (the implementation of a high number of these devices is needed for multi-conjugated adaptive optics). We therefore are investigating a manufacturing process for the pyramidal optical components based on the Deep X-ray Lithography (DXRL) technology, using X-ray synchrotron radiation, and exploiting a lithography dedicated beamline already operating at the ELETTRA Synchrotron in Trieste. This method foresees the irradiation of a transparent and amorphous plastic polymer like PMMA (Plexiglass) with a collimated high intensity X-ray beam. The radiation modifies the internal structure of the irradiated polymer that, once immersed into an appropriate chemical bath, permits its dissolution. In this way, by means of a mask and a series of exposures, it is possible to form a four-faces pyramid. After some preliminary tests, a mechanical system has been built to manufacture the plastic pyramids with tight tolerances. The pyramids manufactured with DXRL can be seen as the final product to be implemented in an AO system. To manufacture large numbers of this components (mass production) we are investigating their molding starting from a pyramid template made with DXRL. The manufactured mold, made by electrodeposition of a metal on the template pyramid, could then be used for the production of pyramids by Injection molding or hot embossing.

Shack-Hartmann based wavefront sensors, used to compensate atmospheric turbulence, appear to be less sensitive than cur-vature based wavefront sensors by more than a magnitude. Besides their read noise free APD detectors and different meas-urement principle, the sensitivity of curvature sensors may benefit from the keystone shaped lenslet geometry which opti-mally balances aperture coverage and illumination of individual subapertures. This paper describes the implementation of a keystone shaped lenslet array for the ALFA Shack-Hartmann based AO system. We compare the novel design with hexago-nal shaped lenslets under different operating conditions such as selected modal basis set and number of compensated modes theoretically and in practice.

Until now, only avalanche photodiodes (APD) have been used as the detectors in curvature wavefront sensors in astronomy. This is due to the strict requirements of very short integration time and very low readout noise. In 1999, Beletic et al. invented a new CCD design which should achieve the same performance as APDs but with higher reliability and lower cost. In addition, this CCD has higher quantum efficiency than APD modules and larger dynamic range, eliminating the need for neutral density filters on bright objects. The CCD was designed and fabricated by MIT Lincoln Laboratory in collaboration with ESO and IfA. R. Dorn extensively tested the CCD in laboratory at ESO and proved that it achieves the predicted performance. CFHT is currently implementing this CCD on PUEO, CFHT’s Adaptive Optics system, to assess its performance for the first time in real conditions on the sky for a direct comparison with the current 19 APD detector system. In this article we present the current implementation scheme and discuss the upgrade we foresee for PUEO NUI, a 104-element high-order curvature AO system envisaged to replace the current AO system at Canada-France-Hawaii Telescope.

The objective of the PYRAMIR project is to complement the Calar Alto Adaptive Optics System - ALFA - with a new pyramid wavefront sensor working in the near IR, replacing the previous tip-tilt tracker arm. Here we describe the Science as well as the Technical motivation for such a system. The optical design will be presented, discussing the particular requirements posed by sensing the wavefronts in the infrared like a cooling system for the opto-mechanical components, etc. We will also talk about the components, like the IR detector we plan to use - PICNIC, as one option, the sucessor of NICMOS3 from Rockwell, together with the AO-Multiplexer. It is described how we expect to integrate the system into the optical, machanical, electronical and control architecture of ALFA.

In the pyramid wavefront sensor some dynamic range is accomplished by modulating the optical signal across the four faces of the pyramid before the dissection and detection of the light. Although this can be realized in different ways, including systems which do not require any moving part, we question and discuss the real needs for such a modulation. In fact, when the closed-loop performance is not perfect, some residual errors on the wavefront sensor are expected and one should take care to allow for enough dynamic range to get a linear response within such a residual range. However, the non-corrected aberrations themselves can be considered as a form of modulation. Higher order uncompensated residuals are equivalent to a modulation for the lower compensated modes. We present a preliminary study showing that this sort of 'natural' modulation could be, at least under certain conditions, enough to reach comparable results with respect to dynamical modulation during correction, hence rising the question of the need of a modulation in the realization of the pyramid wavefront sensor.

A demonstrator of the multi-conjugate adaptive optics concept is under construction at ESO and will be installed on the Nasmyth focus of the VLT. This demonstrator called MAD will have two different wavefront sensor channels: Shack-Hartmann and Layer-Oriented; in this article we only describe the Layer-Oriented one. The Layer-Oriented wavefront sensor can select eight reference stars in the two arc-minutes corrected field of view in order to have a maximum of two references in each quarter of the field. XY stages will remotely adjust the position of each reference star selector. The starlight will be fed onto two detectors and two completely independent loops will drive the deformable mirrors, one conjugated to the ground and the other to an altitude of approximately 8 km. The Layer-Oriented wavefront sensor will use the same CCDs than the Shack-Hartmann channel and the pupil will be divided into 9×9 subapertures both for the high and for the ground layer. The spatial sampling of the subapertures will be different for the two CCDs and their integration time will be tuned to typical values of the conjugated altitudes characteristic wind speed. The overall status of the instrument with respect to optics, mechanics, electronics and software is given hereafter. We also summarize the progress on the procurement phase and give the time schedule for the assembling, integration and testing phases.

We describe some novel technical approaches to implement multi-pyramid wavefront sensing, partially extendable to single pyramid and to any MCAO system. First we introduce an achromatic version of the pyramid, which allows for a much better spatial resolution on the pupil and also relaxes the specifications in term of turned edges. Another item that we discuss is the distribution of tolerances in a layer-oriented AO system which makes attractive, at least in some cases, the usage of pairs of lenslet arrays, leaving only the pyramids free to move over the Field of View, hence relaxing the requirements in terms of roll and yaw in their positioning. Then we discuss the effect of pupil distortion occurring in the layers above the ground during the open loop phases of a MCAO system. Finally we discuss a possible usage of the Modulation Transfer Function as a valuable tool to estimate the correction of a certain Zernike polynomial, achievable with a pyramid wavefront sensor. These items are sketched along with a status of their practical implementation and possible future extensions.

Multi-Conjugate Adaptive Optics (MCAO) is working on the principle to perform wide field of view atmospheric turbulence correction using many Guide Stars located in and/or surrounding the observed target. The vertical distribution of the atmospheric turbulence is reconstructed by observing several guide stars and the correction is applied by some deformable mirrors optically conjugated at different altitudes above the telescope. The European Southern Observatory together with external research institutions is going to build a Multi-Conjugate Adaptive Optics Demonstrator (MAD) to perform wide field of view adaptive optics correction. The aim of MAD is to demonstrate on the sky the feasibility of the MCAO technique and to evaluate all the critical aspects in building such kind of instrument in the framework of both the 2nd generation VLT instrumentation and the 100-m telescope OWL. In this paper we present the conceptual design of the MAD module that will be installed at one of the VLT unit telescope in Paranal to perform on-sky observations. MAD is based on a two deformable mirrors correction system and on two multi-reference wavefront sensors capable to observe simultaneously some pre-selected configurations of Natural Guide Stars. MAD is expected to correct up to 2 arcmin field of view in K band.

The European Southern Observatory (ESO) and the Max Planck Institut fur extraterrestrische Physik (MPE) are jointly developing SINFONI, an Adaptive Optics (AO) assisted Near Infrared Integral Field Spectrometer, which will be installed in the first quarter of 2004 at the Cassegrain focus of YEPUN (VLT UT4). The Adaptive Optics Module, a clone of MACAO, designed and built by ESO, is based on a 60 elements curvature system. It feeds the 3D spectrograph, SPIFFI, designed and built by MPE, with higher than 50% K band Strehl for bright (V<12) on-axis Natural Guide Stars (NGS) and less than 35 mas/hour image motion. The AO-Module will be the first curvature AO system operated in Laser Guide Star (LGS) mode, using a STRAP system for the tip/tilt sensing. The Strehl performance in the LGS mode is expected to be better than 30% in K band.

The Lick Observatory laser guide star adaptive optics system has undergone continual improvement and testing as it is being integrated as a facility science instrument on the Shane 3 meter telescope. Both Natural Guide Star (NGS) and Laser Guide Star (LGS) modes are now used in science observing programs. We report on system performance results as derived from data taken on both science and engineering nights and also describe the newly developed on-line techniques for seeing and system performance characterization. We also describe the future enhancements to the Lick system that will enable additional science goals such as long-exposure spectroscopy.

A rationale is presented for the use of a relatively low-altitude (15km) Rayleigh Laser Guide Star to provide partial adaptive optics correction across a large fraction of the sky on the 4.2m William Herschel Telescope. The scientific motivation in relation to the available instrumentation suite is discussed and supported by model performance calculations, based on observed atmospheric turbulence distributions at the site. The proposed implementation takes the form of a laser system, beam diagnostics, tip-tilt mirror and beacon launch telescope, together with a range-gated wavefront sensor and processing system. It is designed to operate in conjunction with the telescope’s existing facility-class natural guide star AO system, NAOMI. Aspects of the proposed implementation are described as well as the technical features related to the system model and the error budget. In a separate paper the NAOMI AO system itself is presented. Other papers describe a demonstrator system and preliminary Rayleigh beacon wavefront sensing measurements at the site.

This paper describes the upgrades to the Keck II Adaptive Optics (K2 AO) system needed for laser guide star observing. The upgrade, including integration with the laser, is scheduled for completion in the winter of 2003. This upgrade includes the addition of a Low Bandwidth Wavefront Sensor (LBWFS) measuring focus and higher order terms, and a Lawrence Livermore National Lab quad-lens avalanche photodiode detector which monitors tip/tilt. Both observe a dim natural guide star. LBWFS corrections are applied as corrections to the high bandwidth wavefront sensor, which is observing the laser beacon. These subsystems drive focus stages, a deformable mirror, a tip/tilt mirror for the incoming starlight, and a tip/tilt mirror for pointing the propagating laser beam. Taken together, and in concert with the rest of the components of the K2 AO system, they provide the tools and the means to observe the universe as never before.

We report on the ongoing VLT Laser Guide Star Facility project, which will allow the ESO UT4 telescope to produce an artificial reference star for the Adaptive Optics systems NAOS-CONICA and SINFONI. A custom developed dye laser producing >10W CW at 589nm is installed on-board of the UT4 telescope, then relayed by means of a single mode optical fiber behind the secondary mirror, where a 500mm diameter lightweight, f/1 launch telescope is projecting the laser beam at 90 km altitude. We described the design tradeoffs and provide some details of the chosen subsystems. This paper is an update including subsystems results, to be read together with our previous paper on LGSF design description.

For the successful operation of laser referenced adaptive optic systems very powerful lasers for the creation of sodium guide stars are necessary. Here we introduce the design of PARSEC, the cw sodium-line laser for the VLT, and present out first laboratory results on the performance of the system. So far we have achieved a stable output power of 12.8W in a single spatial mode and a single frequency.

One of the critical design drivers for the PARSEC laser was that by the time it is integrated into the VLT Laser Guide Star Facility in 2003 it should be able to run remotely without any hands-on tuning for a period of at least one week. In this contribution we describe some of the methods we have employed to achieve this and take a brief look at the proposed operating procedure.

It is foreseen that Extremly large telescope (ELT) will include Adaptive Optics Systems to provide diffraction-limited images in the near infrared. The preliminary atmospheric turbulence properties which affects laser guide star (LGS) generation are presented here. The results of modern model developed in optical communication through atmosphere is used to evaluate LGS size for many different atmospheric situation, with the turbulence intensity concentrate to different heights. This introduce the focusing problem and requirements above the observatory where the adaptive optics system with a reference source will be install. In other case, focus a laser in the mesosphere is impossible in a generic place. In high altitude observatories this is possible, like in canarian observatories. We took 6 experimental turbulence profile to evaluate the beam radius with altitude. A mean value of 0.2 m are found for the diameter of mesospheric volume illuminated by the laser. This is small enough to create an useful LGS for adaptive optics proposes.

We report on our research to develop a fiber Raman laser for multiple sodium laser guide star adaptive optics. Our goal is a fiber laser source at 589 nm with a diffraction-limited continuous-wave output power of at least 10 W and a linewidth of less than 3 GHz. In this paper we present and model our concept of a frequency-doubled fiber Raman amplifier. Recent advances in commercially available pump and seed laser technologies are making the amplifier approach a highly attractive solution for a turn-key sodium-guide star laser system.

The Gemini Laser Launch Telescope will reside behind the secondary support structure of the Gemini 8m telescope, where it will expand an incoming sodium laser beam to 450 mm diameter and launch it into the sky, co-axial to the main telescope. The tight space and stringent performance specifications have required some innovative approaches in optical and mechanical design.

We present the design, laboratory tests and preliminary field tests of a dynamic refocus system for 351nm Rayleigh beacon laser guide stars. The purpose of dynamic refocus is to increase the beacon signal from a pulsed laser, by maintaining focus in a fixed plane while the laser pulse travels through the atmosphere over an extended height range. The focusing element in our system is a moving concave mirror. The optics have been designed and built to focus on a ring of 5 beacons at 1 arc minute radius at the 6.5 m MMT, covering the range 18 through 40 km. Laboratory tests of image quality resulted in 0.5 arcsec refocused images corresponding to the height range 22 through 28 km, free from spherical aberration. Preliminary field tests were performed on the Mt. Bigelow Observatory 1.5 m telescope, with a frequency tripled, Q switched YLF laser beam projected from a 25 cm telescope. To simulate an off axis sub aperture of the MMT, the laser and telescope axes were set 3 m apart and reimaging optics were placed ahead of the refocus unit to image at the same plate scale as the MMT (500 μm/arcsec). Returns from different heights were selected by gating the detector with a Pockels cell. Returns over a 10 km height range from 8km to 18km were brought into focus for a total mirror motion measured to be 900 μm. The system is now ready for testing dynamic refocus, which will be accomplished by attaching the mirror to a metal resonator tuned to the laser pulse frequency. The range from 23 to 35 km to be used will require a motion of 500 μm.

We briefly recall the principle of the polychromatic laser guide star, which aims at providing measurements of the tilt of incoming wavefronts with a 100% sky coverage, We describe the main results of the feasibility study of this concept undertaken within the ELP-OA porgramme. We finally summarize our plans for a full demonstrator at Observatoire de Haute-Provence.

The idea of achieving Adaptive Optics over the majority of the sky using sodium laser guide stars is reaching maturity on Mauna Kea. However, Mauna Kea is a shared astronomical site with 13 institutions and 11 telescopes. Coordination between observatories with laser guide stars and facilities without laser guide stars must be accomplished to prevent sodium light (Rayleigh scatter and the laser guide star itself) from interfering with science observations at the non-laser facilities. To achieve this goal, a technical working group was organized with participation from several Mauna Kea observatories to discuss and agree upon an automated system for avoiding laser “beam” collisions with other telescopes. This paper discussed the implementation of a Laser Traffic Control System (LTCS) for Mauna Kea including a brief history of the coordination effort, technical requirements and details surrounding implementation of laser beam avoidance software, critical configuration parameters, algorithmic approaches, test strategies used during deployment, and recommendations based upon experiences to date for others intending to implement similar systems.

We measured the Rayleigh scattered light from the laser beam launched from Keck II telescope. The illumination at the pupil of Subaru telescope was recorded using the APDs of the Subaru AO system wavefront sensor. Since the field of view of our wavefront sensor is only 5 arcsec in diameter, it was possible to follow the laser beam as it travels across the telescope pupil. The maximum surface brightness of the laser beam is estimated at 16.2 mag/arcsec².

An important requirement for the VLT Laser Guide Star Facility is the ability to measure the centroid height of the atmospheric sodium layer, located at an altitude of about 90km, to an accuracy of 200m. In this paper, we describe the sodium profiler system designed to achieve this and the expected performance.

In the near future several astronomical observatories in Chile are planning to use sodium laser guide stars to increase the sky coverage provided by their adaptive optics facilities. Knowledge of the mesospheric sodium layer behavior is crucial to predict the performance of future laser guide star adaptive optics systems. Whereas the sodium layer has been observed quite extensively at several locations, many of them in the Northern Hemisphere, very little measurements have been made in Chile. The Gemini Observatory therefore initiated a year-long sodium monitoring campaign at the Cerro Tololo Inter-American Observatory located only a few kilometers away from the Gemini South telescope where a conventional laser guide star facility will be offered to the community in 2005, soon to be upgraded to a multi-conjugate adaptive optics system with five laser guide stars. This paper reports on the laser-based sodium monitoring experimental set up and data reduction techniques, and presents some preliminary results on the sodium column density and layer altitude variations observed from February 2001 to February 2002. Implications for the Gemini South Adaptive Optics system expected performance are presented as well.

The realtime knowledge of the altitude and optical characteristics of a generated Laser Guide Star may be necessary for an independent control of the performances of the associated adaptive optics system. For this ourpose a robotic laser guide star monitoring facility, based on a small telescope physically separated from the observatory has been studied, as a subsystem of the VLT LGS project. After reviewing the state of the art on sodium layer dynamical characteristics, we describe the rationale of the project and the main technical choices. Some particular aspects such as differential field motion and laser plume centroiding algorithms are reviewed in detail. The data processing architecture and the associated pipeline is described, and the expected performances are estimated.

An experimental Adaptive Optics system utilizing a low-altitude Rayleigh Laser Guide Star is presented. The current status of the project is described together with plans for future work, with the ultimate goal of providing demonstration Rayleigh LGS AO science on the 4.2m William Herschel Telescope. The implementation of the demonstration system is described in detail as are the technical challenges encountered during the development of the current shared-launch based system. Other papers in this conference describe a proposal for a facility class LGS system for the WHT and the performance of the observatory's NAOMI adaptive optics system.

The multi-conjugate adaptive optics (MCAO) technique allows to correct the vertical distribution of atmospheric turbulence, extending the isoplanatic angle and the sky coverage. In the Layer-Oriented (LO) approach, each wavefront sensor (WFS) is conjugated to a given atmospheric layer and is linked in closed loop to a deformable mirror, conjugated to the same height. This technique benefits from the co-adding of light coming from the guide stars which can be arbitrarily numerous and faint. In the Multiple Field of View (MFoV) scheme, each LO WFS is looking through a different field-of-view (FoV) in order to extend even further the sky coverage by increasing the photon density in the various layers. With the help of the Software Package CAOS, upgraded with numerical tools dedicated to MCAO and LO, we simulate the whole wavefront sensing process of a LO system based on pyramid sensors. We describe the modelization of the phenomena occurring during the sensing, such as diffraction effects, and detector noise. Furthermore we also use an independent ad hoc code in order to analyze the performance of the MFoV scheme. The performances are given considering realistic natural guide star (NGS) configurations.

In order to achieve moderate Field of View (2 arcmin in diameter) and nearly diffraction limited capabilities, at the reddest portion of the visible spectrum in the interferometric mode of LBT, two sophisticated MCAO channels are required. These are being designed to perform a detailed correction of the atmospheric turbulence through three deformable mirrors per telescope arm: the secondary adaptive mirror and two commercial piezostack mirrors, leading to an overall number of degree of freedom totaling ~ 3000. A combination of numerical and optical coaddition of light collected from natural reference stars located inside the scientific Field of View and in an annular region, partially vignetted, and extending up to ≈ 6 arcmin in diameter, allows for such a performance with individual loops characterized by a much smaller number of degree of freedom, making the real-time computation, although still challenging, to more reasonable levels. We implement in the MCAO channel the dual Field of View layer-oriented approach using natural guide stars, only allowing for limited, but significant, sky coverage.

We present an overview of the optical setup and control algorithms for the multi-conjugate adaptive optics (MCAO) system of the 70cm German Vacuum Tower Telescope (VTT), Observatorio del Teide, Tenerife. The system is designed to remove the strong differential tip/tilt of the present AO system across a field of 30 arcseconds at visible wavelengths. It will consist of two Shack-Hartmann wavefront sensors (WFS) and two deformable mirrors (DM) plus a separate Tip/Tilt mirror. Both wavefront sensors will be situated in the pupil plane of the telescope. One determines the high order wavefront aberrations for the center of the field of view (FOV), the other measures only low order wavefront aberrations, but covers a large FOV in each subaperture. A 35 actuator bimorph mirror and a 37 actuator membrane mirror will correct the ground layer and the tropopause, respectively. For wavefront reconstruction, the mirror eigenmodes will be used. The system will have first light in the first quarter of 2003. Scientific operation is expected to start in April 2003 or July 2003.

The vertical distribution of the turbulence limits the field of view of classical adaptive optics due to the anisoplanatism. Multiconjugate adaptive optics (MCAO) uses several deformable mirrors conjugated to different layers in the atmosphere to overcome this effect. In the last few years, many studies and developments have been done regarding the analysis of the turbulence volume, and the choice of the wavefront reconstruction techniques.An extensive study of MCAO modelisation and performance estimation has been done at OAA and ONERA. The developed Monte Carlo codes allow to simulate and investigate many aspects: comparison of turbulence analysis strategies (tomography or layer oriented) and comparison of different reconstruction approaches. For instance in the layer oriented approach, the control for a given deformable mirror can be either deduced from the whole set of wavefront sensor measurements or only using the associated wavefront sensor. Numerical simulations are presented showing the advantages and disadvantages of these different options for several cases depending on the number, geometry and magnitude of the guide stars.

Layer Oriented MCAO is a promising technique attempting to perform wide field of view correction with Natural Guide Stars. In the extended concept of Multiple Field of View Layer Oriented MCAO the wavefront sensor field of view is enlarged to collect more photons from more Natural Guide Stars and, in principle, significant sky coverages at any galactic latitude are achieved using only Natural Guide Stars. In this paper we address the problem of finding the best magnitude range for the Natural Guide Stars in order to achieve the best correction with the largest sky coverage in the Layer Oriented Multiple Field of View. For a given Field of View and sky direction we consider only the Natural Guide Stars within a given brightness range and we compute the equivalent integrated magnitude. Then we correlate the contribution in sky coverage of the previously considered Natural Guide Stars and we extrapolate which is the magnitude class giving the largest sky coverage. Once identified the more suitable Natural Guide Star magnitude class we discuss the possible implications in the design of a Multiple Field of View Layer Oriented wavefront sensor and we give the order of magnitude for the main parameters, i.e., maximum number of Natural Guide Stars and detector characteristics.

A single-conjugate adaptive optics (AO) system for the 6.5 m MMT based on an adaptive secondary mirror is now undergoing commissioning at the telescope. In addition to providing the basis for a program of scientific observations, the system will serve as a platform for the development of a dual-conjugate AO system in which a constellation of five refocused Rayleigh laser guide stars (RLGS) will provide the majority of the wavefront information. A SCIDAR campaign at a telescope close to the MMT site is beginning to generate explicit measurements of the atmospheric turbulence profile. In this paper, we describe the current system design, and present the results of initial performance simulations using the measured atmospheric parameters.

In order to get first-hand results in the laboratory and possibly on the sky with the Layer-Oriented approach we designed, built, and tested a bread-board for this type of wavefront sensor. This device consists of a single wavefront sensor able to look simultaneously at four references. The positioning of three of the four reference stars with respect to the central one is made by the means of manually adjustable positioning units. A few additional degrees of freedom have been intentionally placed in the system in a way to test the sensitivity of the unit to misplacement and/or misalignment of some optical components. The laboratory set-up includes a crude system to mimic a telecentric F/32 focal plane illuminated by a number of fiber sources that can be placed in several different configurations. Wavefront deformation can be accomplished by placing some fixed deformating plates optically conjugated to several altitudes on the atmosphere. The system is designed in a way to be easily fitted to the existing AdOpt@TNG system, allowing for multiple references, one DM, closed loop operations. Any preliminar result from this activity will be reported. Laboratory experiments includes checking of the theoretical predictions, especially the effectiveness in sensing up to a certain spatial frequency the layers not specifically conjugated to the detector. The results of a demonstrative experiment, showing how the wavefront sensor is able to disentangle layers contribution, are also reported.

The ESO-MAD (MCAO Technological Demonstrator) is a fast track project aimed at demonstrating the maturity of the Multi-Conjugate Adaptive Optics concept through a prototype MCAO instrument that uses only natural guide stars. This prototype features two different wavefront sensing architectures (Shack-Hartmann and Layer-Oriented), two deformable mirrors and one tip/tilt stage. One of the objectives of MAD is also to explore computing architectures different from the ones adopted so far. MAD-RTC is based on the latest generation general-purpose processors in a parallel architecture that can easily (even though not unexpensively) scale up to accomodate large or very large MCAO systems. MAD-RTC is a multi-wavefront-sensor multi-algorithm real time computer implemented in a Quad-G4 PPC computing board. It is designed to be a test-bed to study different solutions for the future MCAO systems: it can accept data from multiple CCDs in different configurations, use different reconstruction and control algorithms and drive multiple mirrors simultaneously.

Multiconjugate adaptive optics employing several deformable mirrors conjugated at different altitudes has been proposed in order to extend the size of the corrected field of view [FOV] with respect to the size of the corrected FOV given by a classical adaptive optics system. A three dimensional measurement of the turbulent volume is needed in order to collect the information to command the several deformable mirrors. Given a set of guide stars in the field of view, this can be done both using tomography, in which several wavefront sensors are used, each of them coupled to one of the guide stars, or layer oriented techniques, in which wavefront sensors are coupled to a given layer in the atmosphere, and collect light from the whole set of guide stars. We will call this type of measurements optical layer oriented. This type of measurements can be also obtained combining in a numerical way, tomographic measurements. This hybrid approach is called numerical layer oriented. In order to compare their performance, we present an analytical study of the signal to noise ratio [SNR] in the measurements for the two techniques. Optical layer oriented is shown to be more efficient in the range of faint flux and large number of guide stars, while low detector noise will allow numerical layer oriented schemes to be more efficient in terms of SNR.

In this paper, we review the salient facts for a range of available atmosphere emulation technologies, and in the framework of the ESO Multi-Conjugate-AO demonstrator project, aptly called MAD, we present our phase screen test results for silver-sodium ion-exchange, transmissive phase screens. We find (a) that the measured power spectrum of phase fluctuations is consistent with the input Von Karman spectrum and (b) that by tracking the best focus of ten spots formed by a silver-sodium ion-exchange micro-lens array, it was found that the wavelength dependence of 1.266μm of phase-shift is 1.5±2.5% relative to air in the wavelength range 550nm to 800μm. Additionally, we present our optical design and specifications for MAPS, the Multi-Atmospheric Phase screens and Stars instrument that will be used to test MAD before shipment to the VLT. It includes glass screens conjugate to the 0.25km, 3.0km, and 9.0km atmospheric layers above the telescope. We explain the reasoning behind the choice of pupil size and implications for phase screen proximity, footprint sizes, and wind speed gradients. Our design mimics the VLT Nasmyth F/15 focal plane in terms of plate scale, field of view, high Strehl, and field curvature.

We present a progress report of the solar adaptive optics (AO) development program at the National Solar Observatory (NSO) and the Big Bear Solar Observatory (BBSO). Examples of diffraction-limited observations obtained with the NSO low-order solar adaptive optics system at the Dunn Solar Telescope (DST) are presented. The design of the high order adaptive optics systems that will be deployed at the DST and the BBSO is discussed. The high order systems will provide diffraction-limited observations of the Sun in median seeing conditions at both sites.

NAOMI (Nasmyth Adaptive Optics for Multi-purpose Instrumentation) is a recently completed and commissioned astronomical facility on the 4.2m William Herschel Telescope. The system is designed to work initially with Natural Guide Stars and also to be upgradeable for use with a single laser guide star. It has been designed to work with both near infrared and optical instrumentation (both imagers and spectrographs). The system uses a linearised segmented adaptive mirror and dual-CCD Shack-Hartmann wavefront sensor together with a multiple-DSP real-time processing system. Control system parameters can be updated on-the-fly by monitoring processes and the system can self-optimize its base optical figure to compensate for the optical characteristics of attached scientific instrumentation. The scientific motivation, consequent specification and implementation of NAOMI are described, together with example performance data and information on future upgrades and instrumentation.

Rethinking the efficient use of 4m-class telescopes in the dawning era of larger facilities is a timely but challenging debate. The extensive use of PUEO for imaging (and now spectroscopy) has kept CFHT at the forefront of scientific research with adaptive optics since its commissioning in 1996. Even though larger facilities are now starting to think about ways of implementing high order AO systems, we believe the medium size of the CFHT and the excellent quality of our site on Mauna Kea is a perfect combination to reach the highest performances with a high order AO system. The fields of application of high order adaptive optics are exciting: They include extremely high contrast imaging and coronography in the near-infrared and diffraction-limited imaging in the optical, with the corresponding gain in angular resolution. Specific science examples are described in and adjacent paper (Menard et al, these proceedings (4839-133)), and planned instrumentation in the form of four quadrant coronograph or existing dual (or triple) wavelength imagers (such as TRIDENT) would benefit tremendously from >90% Strehl ratios in the K band. Simulations of a high order (104 electrodes) curvature system have been performed and produce the required performance and are presented in an adjacent paper (Lai & Craven-Bartle, (4860-28)). Technologically, the system is quite simple and re-uses most of the opto-mechanics of the existing PUEO. Deformable mirrors and real time computers are well within existing (and commercially available) specifications. An innovative solution of using a dedicated low read noise CCD camera (specifically for curvature systems) overcomes the potential cost drawbacks of using avalanche photo-diodes (APDs). This detector is described in detail in an adjacent paper (Cuillandre et al, these proceedings (4839-31)).

Pushing the adaptive compensation of turbulence into the visible range remains a challenging task, despite the progress of AO technology. An AO system for SOAR, now under conceptual study, will be able to reach diffraction-limited resolution at 0.5-0.7 microns with natural guide stars as faint as magnitude 12, enabling studies of stellar vicinities for faint companions, nebulosity, etc. During the second stage of the project a Rayleigh laser guide star will be implemented. In this mode, only the lowest turbulent layers will be compensated. The angular resolution will be only two times better than natural seeing, but, in exchange, the uniformly compensated field will reach 2-3 arc-minutes, offering unique capabilities in crowded fields (clusters, nearby galaxies).

A NASA research contract (NAS1-00116) was awarded to Ball Aerospace & Technologies Corp. in January 2000 to study wide field-of-view adaptive optical systems. These systems will be required on future high resolution Earth remote sensing systems that employ large, flexible, lightweight, deployed primary mirrors. The deformations from these primary mirrors will introduce aberrations into the optical system, which must be removed by corrective optics. For economic reasons, these remote sensing systems must have a large field-of-view (a few degrees). Unlike ground-based adaptive optical systems, which have a negligible field-of-view, the adaptive optics on these space-based remote sensing systems will be required to correct for the deformations in the primary mirror over the entire field-of-view. A new error function, which is an enhancement to conventional adaptive optics, for wide field-of-view optical systems will be introduced. This paper will present the goals of the NASA research project and its progress. The initial phase of this research project is a demonstration of the wide field-of-view adaptive optics theory. A breadboard has been designed and built for this purpose. The design and assembly of the breadboard will be presented, along with the final results for this phase of the research project. Finally, this paper will show the applicability of wide field-of-view adaptive optics to space-based astronomical systems.

The adaptive secondary for the MMT (called MMT336) is the first mirror of its kind. It was designed to allow the application of wavefront corrections (including tip-tilt) directly at the secondary mirror location. Among the advantages of such a choice for adaptive optics operation are higher throughput, lower emissivity, and simpler optical setup. The mirror also has capabilities that are not found in most correctors including internal position feedback, large stroke (to allow chopping) and provision for absolute position calibration. The 336 actuator adaptive secondary for MMT has been used daily for over one year in our adaptive optics testing facility which has built confidence in the mirror operation and allowed us to interface it to the MMT adaptive optics system. Here we present the most recent data acquired in the lab on the mirror performance. By using interferometer measurements we were able to achieve a residual surface error of approximately 40nm rms. Coupling the mirror with a Shack-Hartmann wavefront sensor we obtained a stable closed loop operation with a -3dB closed loop bandwidth of approximately 30Hz limited by the wavefront sensor frame rate. We also present some preliminary results that show a 5Hz, 90% duty cycle, ±5 arcsec chopping of the mirror. Finally the experience gained and the problems encountered during the first light adaptive optics run at the telescope will be briefly summarized. A more extensive report can be found in another paper also presented at this conference.

Future concepts of ultra large space telescopes include the telescopes with segmented silicon mirrors and inflatable polymer mirrors. Primary mirrors for these systems cannot meet optical surface figure requirements and are likely to generate over several microns of wavefront errors. In order to correct for these large wavefront errors, high stroke optical quality deformable mirrors are required. The deformable mirror in this paper consists of an optical quality membrane mirror backed by a piezoelectric unimorph actuator array. A fabrication technology has been recently developed for transferring an entire wafer-level mirror membrane from one substrate to another. A thin membrane, 100 mm in diameter, has been successfully transferred without using adhesives or polymers. The measured peak-to-valley surface figure error of a wafer-level transferred and patterned membrane (1 mm × 1 mm × 0.016 mm) is only 9 nm. The mirror element actuation principle is based on a piezoelectric unimorph. A voltage applied to the piezoelectric layer induces stress in the longitudinal direction causing the film to deform and pull on the mirror connected to it. The advantage of this approach is that the small longitudinal strains obtainable from a piezoelectric material at modest voltages are thus translated into large vertical displacements. Modeling is performed for unimorph membranes consisting of clamped rectangular membranes with piezoelectric layers with variable dimensions. The membrane transfer technology combined with the piezoelectric unimorph actuator concept constitutes the successful demonstration of high quality, compact, large stroke continuous deformable mirror devices, resulting in compact AO systems for use in ultra large space telescopes.

Next generation giant telescopes as well as next generation instrumentation for 10m-class telescopes relies on the availability of highly performing adaptive optical systems. Different types of AO systems are currently under study, including Multi-Conjugate AO (MCAO), high dynamic range AO, and low-order AO for distributed partial correction AO. These systems require a large variety of deformable mirrors with very challenging parameters. The development of new technologies based on micro-opto-electro-mechanical systems (MOEMS) is promising for future deformable mirrors. The major advantages of the micro-deformable mirrors (MDM) are their compactness, scalability, and specific task customization using elementary building blocks. We are currently developing a MDM based on an array of electrostatic actuators with attachments to a continuous mirror on top. A high optical quality mirror is the most challenging building block for this device. The originality of our approach lies in the elaboration of a sacrificial layer and of a structural layer made of polymer materials. With this structure, very efficient planarization has been obtained: the long-distance flatness is below 0.2 μm, the print-through of localized 9μm steps is reduced to below 0.5μm and a rms roughness of 15 nm has been measured over the surface. The integration of this mirror surface on top of an actuator array is under investigation.

Semipassive bimorph mirrors for wavefront correction in imaging systems and high power lasers were developed and investigated. Different types of substrates and active piezoceramics materials were considered to fabricate temperature independent shape of the mirror surface and to maximize the sensitivity of the mirror. High reflectivity coatings for different wavelengths were studied. Bimorph mirrors were used for wavefront correction in living human eye and high-power solid state lasers.

The two adaptive secondary (AS) mirrors for LBT (LBT672) represent the new generation of the AS technology. Their design is based on the experience earned during the extensive tests of the previous generation unit (the MMT AS mirror). Both the mechanics and the electronics have been revised, improving the stability, reliability, maintenance and computational power of the system. The deformable mirror of each unit consists of a 1.6mm-thick Zerodur shell having a diameter of 911mm. The front surface is concave to match the Gregorian design of the telescope. Its figure is controlled by 672 electro-magnetic force actuators that are supported and cooled by an aluminum plate. The actuator forces are controlled using a combination of feed-forward and de-centralized closed loop compensation, thanks to the feedback signals from the 672 co-located capacitive position sensors. The surface reference for the capacitive sensors is a 50mm-thick Zerodur shell faced to the back surface of the thin mirror and rigidly connected to the support plate of the actuators. Digital real-time control and unit monitoring is obtained using new custom-made on-board electronics based on new generation 32bit floating-point DSPs. The total computational power (121 Gflop/s) of the LBT672 units allows using the control electronics as wave-front computer without any reduction of the actuator control capability. We report the details of the new features introduced in the LBT672 design and the preliminary laboratory results obtained on a prototype used to test them. Finally the facility in Arcetri to test the final LBT672 units is presented.

The trend towards ever larger telescopes and more advanced adaptive optics systems is driving the need for deformable mirrors with a large number of low cost actuators. Liquid mirrors have long been recognized a potential low cost alternative to conventional solid mirrors. By using a water or oil based ferrofluid we are able to benefit from a stronger magnetic response than is found in magnetic liquid metal amalgams and avoid the difficulty of passing a uniform current through a liquid. Depositing a thin silver colloid known as a metal liquid like film (MELLF) on the ferrofluid surface solves the problem of low reflectivity of pure ferrofluids. This combination provides a liquid optical surface that can be precisely shaped in a magnetic field. We present experimental results obtained with a prototype deformable liquid mirror based on this combination.

In this paper we apply, recently developed, distributed control techniques to the control of an adaptive secondary mirror. The control technique implemented uses the l2 induced norm as a performance criterion. A finite difference model for the mirror is developed using shift operators. We determine the necessary constants from the physical properties of the mirror, from the typical dimensions of such a system, and from our performance requirements for operating the telescope in the near infrared region.The closed loop system is simulated by modelling the deformable mirror in Abaqus, a commercial finite element software, and by implementing the controller in Matlab.

The continuous thin-plate type with discrete actuators is widely used for active or adaptive mirrors in the medium size range from about 10cm up to 2m in diameters. The performance of a thin-plate deformable mirror could be characterized by the influence function of an actuator and the layout of the actuators. This paper first derives an equation which estimates influence functions of thin-plate deformable mirrors based on the analytic calculation and finite element analysis. Then the performance analysis for the case of equi-spaced actuators is presented.

The Large Binocular Telescope will perform its first level AO correction at visual wavelengths by the two Gregorian secondary mirrors. Each unit is made by a 911 mm diameter and 1.6 mm thick Zerodur shell which shape is controlled by 672 electromagnetic actuators at 1 kHz rate. The shape of each mirror is referred to a Zerodur 50 mm thick backplate through a set of capacitive sensors co-located with the actuators. Each adaptive secondary unit embeds its real time computer for actuator control and communication. Each unit is aligned into the secondary hub by a 6 d.o.f. hexapod system. The construction of the AO units started this year, while the hexapods have been completed in 2001. We present in this paper the final design of the adaptive secondary systems with particular emphasis on the modifications that we made based on the MMT adaptive secondary experience. We will also report the first results of the subsystems development tests.

The adaptive secondary mirror is a fundamental part in the LBT adaptive optics architecture. The thin, continuous mirror is controlled by 672 electromagnetic actuators (voice coil motors) with local position feedback (capacitive sensor) and allows to perform from tip-tilt to high order wavefront correction, but also chopping. The adaptive secondary is controlled by a DSP-based dedicated electronics. The control electronics does not only implement the mirror position control tasks, but does also realize the Real Time Reconstructor (RTR). The control system, while maintaining a similar architecture to the MMT adaptive secondary one, shows a substantial enhancement in terms of computational power, rising in the range of hundreds of Gigaflops. This allows to minimize the computational time required to apply the wavefront correction pattern from the wavefront sensor acquisition, even in case of high order reconstructor dynamics. The electronics is housed in compact cooled crates placed in the adaptive secondary hub. Apart from the power supply lines, it is connected to the other components of the adaptive control system just through a very high speed fiber optic link, capable of 2.9 Gigabit/s of actual data throughput. The control system has been designed according to modular concept, so that the number of channels can be easily increased or reduced for adapting the electronics to different correctors. A substantial effort has been dedicated to the flexibility and on-field configurability of system. In this frame, the same electronics (or part of it) can be easily adapted to become the building block for the data processing unit required for Multi-Conjugated Adaptive Optics

The temporal fluctuations of the optical turbulence at different altitudes in the atmosphere are investigated. About 2400 C²_N profiles, measured with a generalized scidar at the Observatorio Astronomico Nacional de San Pedro Martir, are used to determine the mean amplitude and characteristic time of the relative fluctuations of the optical turbulence in five atmospheric slabs. It is found that the saturation values of the relative fluctuations of the turbulence are 26%, 37%, 28% and 31% and the mean time for reaching 63% of these saturation values are 0.50, 0.20, 0.20 and 0.16 hours for the altitude slabs [2;4], [4;9], [9;16] and [16;21]km, respectively.

The characterization of the optical turbulence (OT) should be considered as one of the most important challenges related to the ELTs feasibility. During the past few years it was proved that the limitations of the adaptive optics techniques due to the turbulence change as the AO techniques evolve. For this reason, it would be useful to develop a flexible tool that could characterize the turbulence in an exhaustive way and that permits, at the same time, relatively simple adaptations in order to satisfy new requirements of the AO techniques and the site testing. In this contribution I will talk about the usefulness of meso-scale atmospherical models for a 3D characterization of the optical turbulence. I will present the principal potentialities of the numerical technique in the astronomical context and the different methods that can be employed to simulate the OT. I will present the progress obtained in OT simulations using the 'Meso-Nh' model on some of the best observatories in the world: Cerro Paranal (Chile), Roque de Los Muchachos (Canary Islands) and San Pedro Martir (Mexico). I will show which is the score of success presently attained with the numerical technique and I will try to sketch how such simulations could be a useful support for AO techniques. Finally, I will talk about applications of the model to the selection of astronomical sites for the ELTs.

The wavefront sensors of adaptive optics systems of astronomical telescopes collect an abundance of high temporal resolution information about the distortions that are introduced to the incoming wavefront by atmospheric turbulence. Although this information can theoretically be used to analyze the turbulence conditions above the telescope at the given time, it is often discarded. The reason for this dismissal of seemingly useful information is usually the difficulty of separating atmospheric and instrumental contributions to the wavefront sensor measurements and thus of obtaining reliable estimates of the atmospheric turbulence conditions. In this paper we describe an effort to overcome these problems for wavefront sensor measurements taken by the Keck telescopes on Mauna Kea. We discuss different methods of deriving turbulence parameters, such as coherence length and time and the outer scale of turbulence, and present first results.

We present the design of and recent results from the Large Binocular Telescope (LBT) facility SCIDAR. To our knowledge, this work will produce the first SCIDAR designed as a user instrument for routine seeing measurements in support of telescope operations. Using a commercial, off-the-shelf approach, we have minimized the resources required for system construction.

The MASS (Multi-Aperture Scintillation Sensor) instrument consists of a 14-cm off-axis reflecting telescope and a detector unit which measures the scintillations of single stars in four concentric zones of the telescope pupil using photo-multipliers. Statistical analysis of these signals yields information of the vertical turbulence profile with a resolution of dh/h=0.5. We describe the instrument and present the results of its first field tests, including comparisons with DIMM seeing monitor and generalized SCIDAR. MASS will be used to obtain the extensive statistics of turbulence profiles at potential sites of future giant telescopes, as needed to predict the quality of adaptive seeing compensation.

A prototype of correlation tracker (CT) designed for the Space Solar Telescope (SST) has been realized in National Astronomical Observatories of CAS. We designed a special optical system for the development and test of the CT system. We estimate the image motion by cross-correlation method via FFT and take full advantage of the real-valued FFT to reduce the amount of calculation. For a possible realization of the processing unit used in space, we use a "1 bit correlation" method, which will greatly simplify the hardware implementation of FFT. A resolution of 0.01" has been obtained corresponding to the object space of the telescope, with the -3dB closed-loop error cutoff frequency of 38Hz along x axis and 40Hz along y-axis achieved.

The Adaptive Optics for the Telescopio Nazionale Galileo module (namely AdOpt@TNG) implements the pyramid wavefront sensor as a unique feature. This allows to get valuable information on its performance on the sky. An updated overview of the results obtained so far is shown, including a discussion on the sources of errors in the closed loop operation, distinguishing them between the ones specific of the pyramid wavefront sensor and the one more related to the system as a whole. This system allows also for a number of experiments and check of the sensitivity of such a wavefront sensor, especially in comparison with other types of sensing units. The ways to accomplish such an experiment in a convincing way are shown along with the first results obtained so far. Finally, we describe how and up to which extent a number of practical problems encountered in the near past can be solved implementing the recent new ideas on the pyramid theme, many of which popped up from our "lessons learned".

We propose in this paper an optimal closed loop control law for multiconjugate adaptive optics (MCAO), based on a Kalman filter and a feedback control. The so-called open loop optimal phase reconstruction is recalled. It is based on a Maximum A Posteriori (MAP) approach. This approach takes into account wavefront sensing noise and also makes use of a turbulence profile model and Kolmogorov statistics. We propose a closed-loop modelization via a state-space representation. A Kalman filter is used for the phase reconstruction. This approach is a closed loop generalization of the MAP open loop estimator. It uses the same spatial prior in addition with a temporal model of the turbulence. Results are compared with the Optimized Modal Gain Integrator approach in the classical adaptive optics case and in an MCAO-like case.

A main objective of adaptive optics is to maximize closed-loop Strehl, or, equivalently, minimize the statistical mean-square wavefront residual. Most currently implemented AO wavefront reconstructors and closed-loop control laws do not take into account either the correlation of the Kolmogorov wavefronts over time or the modified statistics of the residual wavefront in closed loop. There have been a number of attempts in the past to generate "predictive" controllers, which utilize wind speed and Cn2 profiles and incorporate one or two previous time steps. We present here a general framework for a dynamic controller/reconstructor design where the goal is to maximize mean closed-loop Strehl ratio over time using all previous data and exploiting the spatial-temporal statistics of the Kolmogorov turbulence and measurement noise.

We consider the use of pupil masks with AO-compensated telescopes, calculating fringe energy SNR for a variety of object brightness values and for different degrees of AO compensation. We propose a way to specify the optimum subaperture diameter for an N-subaperture mask and argue for the use of masks, in combination with full-aperture data, to increase SNR at the highest spatial frequencies the telescope can sample. Use of full-aperture data in combination with high-SNR masked-aperture data can yield imagery with higher spatial resolution than full-aperture data alone.

Adaptive optics (AO) systems currently under investigation will require at least two orders of magitude increase in the number of actuators, which in turn translates to effectively a 10⁴ increase in compute latency. Since the performance of an AO system invariably improves as the compute latency decreases, it is important to study how today's computer systems will scale to address this expected increase in actuator utilization. This paper answers this question by characterizing the performance of a single deformable mirror (DM) Shack-Hartmann natural guide star AO system implemented on the present-generation digital signal processor (DSP) TMS320C6701 from Texas Instruments. We derive the compute latency of such a system in terms of a few basic parameters, such as the number of DM actuators, the number of data channels used to read out the camera pixels, the number of DSPs, the available memory bandwidth, as well as the inter-processor communication (IPC) bandwidth and the pixel transfer rate. We show how the results would scale for future systems that utilizes multiple DMs and guide stars. We demonstrate that the principal performance bottleneck of such a system is the available memory bandwidth of the processors and to lesser extent the IPC bandwidth. This paper concludes with suggestions for mitigating this bottleneck.

Extensive system modeling and analysis clearly shows that system latency is a primary performance driver in closed loop adaptive optical systems. With careful attention to all sensing, processing, and controlling components, system latency can be significantly reduced. Upgrades to the Starfire Optical Range (SOR) 3.5-meter telescope facility adaptive optical system have resulted in a reduction in overall latency from 660 μsec to 297 μsec. Future efforts will reduce the system latency even more to the 170 msec range. The changes improve system bandwidth significantly by reducing the "age" of the correction that is applied to the deformable mirror. Latency reductions have been achieved by increasing the pixel readout pattern and rate on the wavefront sensor, utilizing a new high-speed field programmable gate array (FPGA) based wavefront processor, doubling the processing rate of the real-time reconstructor, and streamlining the operation of the deformable mirror drivers.

Many reconstructors, or Real Time Controllers (RTC), for mono-conjugate AO systems are currently operating with many more about to be commissioned. The advent of faster and more efficient CPUs has permitted this task to be accomplished on a single processing element, for all but the highest order systems. However, the demands on the RTC increase by an order of magnitude or so in the case of a Multi-Conjugate AO (MCAO) system. Multiple Wavefront Sensors (WFS) and multiple deformable mirrors increase the complexity, processing load and data flow rates that the RTC must deal with. No currently available single processing unit is capable of meeting this demand and retain the advantages of a cost-effective, flexible system. Multiple processing units must be employed. We present in this paper a general architecture that addresses these issues. We present an analysis of the requirements of the Gemini South MCAO system on the RTC. This is followed by an algorithmic decomposition that simplifies the problem, lending itself to the use of commercially available multi-CPU single board computers. This is supported by the results of benchmark tests aimed at verifying the capabilities of one sample SBC. We conclude by presenting a description of the extendability of this architectural approach in the face of yet higher demands such as more mirrors, WFSs or complexity.

Subaru AO a 36 element Curvature Sensor was developed for the 8.2m Subaru Telescope operated by the National Astronomical Observatory of Japan. The system is already over one year on sky and now offered for open use observation. During the time from first light to open use we did major changes in software to operate the instrument more easy and efficient and also improved the performance by tuning the control loop. We will discuss the software model and show what we learned and how we managed the crucial points of this implementation. After this we will have a look at how we can improve the system further.

Altair is the adaptive optics system developed by NRC Canada for the Gemini North Telescope. Altair uses modal control and a quad-cell based Shack-Hartmann wavefront sensor. In order for Altair to adapt to changes in the observing conditions, two optimizers are activated when the AO loop is closed. These optimizers are the modal gain optimizer (MGO) and the centroid gain optimizer (CGO). This paper discusses the implementation and timing results of these optimizers.

A major difficulty with wavefront slope sensors is their insensitivity to certain phase aberration patterns, the classic example being the waffle pattern in the Fried sampling geometry. As the number of degrees of freedom in AO systems grows larger, the possibility of troublesome waffle-like behavior over localized portions of the aperture is becoming evident. Reconstructor matrices have associated with them, either explicitly or implicitly, an orthogonal mode space over which they operate, called the singular mode space. If not properly preconditioned, the reconstructor's mode set can consist almost entirely of modes that each have some localized waffle-like behavior. In this paper we analyze the behavior of least-squares reconstructors with regard to their mode spaces. We introduce a new technique that is successful in producing a mode space that segregates the waffle-like behavior into a few "high order" modes, which can then be projected out of the reconstructor matrix. This technique can be adapted so as to remove any specific modes that are undesirable in the final recontructor (such as piston, tip, and tilt for example) as well as suppress (the more nebulously defined) localized waffle behavior.

Altair is the AO system developed by the National Research Council of Canada for the Gemini North Telescope. Currently in the integration phase, the performance of the system, and in particular, of the real-time controller (RTC), is being fully characterised. We begin with a review of the requirements for the RTC, the context within the rest of the Wavefront Control System, and a description of the processing the RTC is responsible for. We then present the actual performance results of the RTC integrated within Altair. We compare these with the benchmarked results.

Multi-conjugate adaptive optical (MCAO) systems with from 10,000 to 100,000 degrees of freedom have been proposed for future giant telescopes. Using standard matrix methods to compute, optimize, and implement wavefront reconstruction algorithms for these systems is impractical, since the number of calculations required to compute (apply) the reconstruction matrix scales as the cube (square) of the number of AO degrees of freedom. Significant improvements in computational efficiency are possible by exploiting the sparse and/or periodic structure of the deformable mirror influence matrices and the atmospheric turbulence covariance matrices appearing in these calculations. In this paper, we review recent progress in developing an iterative sparse matrix implementation of minimum variance wavefront reconstruction for MCAO. The basic method is preconditioned conjugate gradients, using a multigrid preconditioner incorporating a layer-oriented, iterative smoothing operator. We outline the key elements of this approach, including special considerations for laser guide star (LGS) MCAO systems with tilt-removed LGS wavefront measurements and auxiliary full aperture tip/tilt measurements from natural guide stars. Performance predictions for sample natural guide star (NGS) and LGS MCAO systems on 8 and 16 meter class telescopes are also presented.

Waffle modes are known in adaptive optics as wavefront perturbations giving a null or very small measurement on the wavefront sensor; when these modes appear on the deformable mirror, they are allowed to grow arbitrarly, generating loop instability and degrading the correction. In Multi-Conjugate Adaptive Optics, apart from the invisible modes related to the properties of the wavefront sensors, additional waffle modes may arise depending on the spatial configuration of the guide stars. The relevance of the problem to the layer-oriented approach is discussed, along with a practical method to identify them for a given guide stars asterism. Furthermore some strategies are described to face the problem at the real-time software level, with no additional complication on the hardware side.

Multi-conjugate adaptive optics (MCAO) systems with 10⁴-10⁵ degrees of freedom have been proposed for future giant telescopes. Using standard matrix methods to compute, optimize, and implement wavefront control algorithms for these systems is impractical, since the number of calculations required to compute and apply the reconstruction matrix scales respectively with the cube and the square of the number of AO degrees of freedom. In this paper, we develop an iterative sparse matrix implementation of minimum variance wavefront reconstruction for telescope diameters up to 32m with more than 10⁴ actuators. The basic approach is the preconditioned conjugate gradient method, using a multigrid preconditioner incorporating a layer-oriented (block) symmetric Gauss-Seidel iterative smoothing operator. We present open-loop numerical simulation results to illustrate algorithm convergence.

Adaptive optics systems with Shack-Hartmann wavefront sensors require reconstruction of the atmospheric phase error from subaperture slope measurements, with every sensor in the array being used in the computation of each actuator command. This fully populated reconstruction matrix can result in a significant computational burden for adaptive optics systems with large numbers of actuators. A method for generating sparse wavefront reconstruction matrices for adaptive optics is proposed. The method exploits the relevance of nearby subaperture slope measurements for control of an individual actuator, and relies upon the limited extent of the influence function for a zonal deformable mirror. Relying only on nearby sensor information can significantly reduce the calculation time for wavefront reconstruction. In addition, a hierarchic controller is proposed to recover some of the global wavefront information. The performance of these sparse wavefront reconstruction matrices was evaluated in simulation, and tested on the Palomar Adaptive Optics System. This paper presents some initial results from the simulations and experiments.

The performance of Fourier transform (FT) reconstructors in large adaptive optics systems with Shack-Hartmann sensors and a deformable mirror is analyzed. FT methods, which are derived for point-based geometries, are adapted for use on continuous systems. Analysis and simulation show how to compensate for effects such as misalignment of the deformable mirror and wavefront sensor gain. Further filtering methods to reduce noise and improve performance are presented. These modifications can be implemented at the filtering stage, preserving the speed of FT reconstruction and providing flexibility by allowing on-the-fly filter adaptation. Simulation of a large system shows how compensated FT methods can have equivalent or better performance to slower vector-matrix-multiply methods. The best-performing FT method is the fastest to compute, has lower noise propagation and does not suffer from waffle errors.

Differential Atmospheric Refraction (DAR) reduces image quality on ground-based 10-m telescopes equipped with Adaptive Optics (AO). Particularly affected are the long exposure data taken in narrow-band imaging or spectroscopic mode. The magnitude of the DAR is a function of the effective wavelength of the wavefront sensor detector, meteorological variables, the observing wavelength and the elevation of the observation. In this paper, we present the approach taken by the Keck Adaptive Optics team to compensate for DAR during AO observing. This paper will present a description and illustration of the problem and our solution to it, including some implementation details. This paper also presents some tips on observing techniques, along with some details on current performance, a description of the issues limiting the performance, and our plans for the future.

We present resolved images of the occultation of a binary star by Titan, recorded with the Palomar Observatory adaptive optics system on 20 December 2001 UT. These constitute the first resolved observations of a stellar occultation by a small body, and demonstrate several unique capabilities of diffraction-limited imaging systems for the study of planetary atmospheres. Two refracted stellar images are visible on Titan's limb throughout both events, displaying scintillations due to local density variations. Precise relative astrometry of the refracted stellar images with respect to the unnocculted component of the binary allows us to directly measure their altitude in Titan's atmosphere. Their changing positions also lead to simple demonstration of the finite oblateness of surfaces of constant pressure in Titan's mid-latitude stratosphere, consistent with the only previous measurement of Titan's zonal wind field.

Use of the highly sensitive Hokupa'a/Gemini curvature wavefront sensor has allowed, for the first time, direct adaptive optics (AO) guiding on very low mass (VLM) stars with cool spectral types (M8.0-L0.5). These low mass (Mass < 100 M_JUPITER) objects are very cool (T_eff<3000K) and have very low luminosities (V ⪆ 20 at D = 20 pc) but are red enough (V-I ~ 4) for the Houkupa'a curvature WFS to guide on the reddest (λ_eff~ 0.8 μm) photons. This is the only high-resolution (FWHM~0.1") survey ever of stars cooler than M7 from the ground. We guided on 39 such objects and detected 9 VLM binaries (7 of which were discovered for the first time to be binaries). Most of these systems (55%) are tight (separation <5 AU) and have similar masses (Δ Ks<0.8 mag; 0.85<q<1.0). However, 2 systems (LHS 2397a, and 2M2331016-040618) have large Δ Ks>2.38 mag and consist of a VLM star orbited by a much cooler L6.5-L8.5 brown dwarf companion. Based on our initial flux limited (Ks<12 mag) survey of 39 M8.0-L0.5 stars (mainly from the sample of Gizis et al. 2000) we find a binary fraction in the range 19±7% for M8.0-L0.5 binaries with separations >2.6 AU. This is slightly less than the 32±9% measured for more massive (M0-M4) stars over the same separation range (Fischer & Marcy 1992). It appears M8.0-L0.5 binaries (as well as L and T dwarf binaries) have a much smaller semi-major axis distribution peak (~4 AU) compared to more massive M and G stars which have a broad peak at larger ~30 AU separations. We also find no VLM binaries (M_tot<0.18 Msun) with separations >20 AU. We find that a velocity ``kick'' of ~3 km/s can reproduce the observed cut-off in the semi-major axis distribution at ~20 AU. This kick may have been from the VLM system being ejected from its formation mini-cluster.

Deconvolution is a necessary tool for the exploitation of adaptive optics corrected images, because the correction is partial. The Maximum A Posteriori (MAP) framework is used to derive a deconvolution method (MISTRAL) that combines the data with our knowledge of the noise statistics as well as our prior information about the object and the variability of the Point Spread Function. The deconvolution of experimental and scientific data illustrates the capabilities of this method.

Adaptive optics images of Iota Cassiopeia taken in the I and H bands with the 3.63m AEOS telescope on Haleakala show a strong waffle-like pattern. The Parametric Blind Deconvolution (PBD) point spread function is modified to include such an optical aberration pattern, and position angles, separations, and magnitude differences are obtained for the components of the quadruple star system. The H band image of the astrometric companion is the first ever, the I band image is the first to show all four components, and by combining our data with previous visual and speckle measurements, the orbits and masses of components A and Aa are derived for the first time. The faint sub-solar mass component is 2.3, 2.9, 3.7, and 4.2 magnitudes fainter than its nearby (0.4") companion at H, I, V, and B bands, respectively.

We describe a coronagraph facility built for use with the 4.2 metre William Herschel Telescope (WHT) and its adaptive optics system (NAOMI). The use of the NAOMI adaptive optics system gives an improved image resolution of ~0.15 arcsec at a wavelength of 2.2 microns. This enables our Optimised Stellar Coronagraph for Adaptive optics (OSCA) to null stellar light with smaller occulting masks and thus allows regions closer to bright astronomical objects to be imaged. OSCA is a fully deployable instrument and when in use leaves the focus of the NAOMI beam unchanged. This enables OSCA to be used in conjunction with a number of instruments already commissioned at the WHT. The main imaging camera to be used with OSCA will be INGRID; a 1024×1024 HgCdTe cooled SWIR detector at the NAOMI focus. OSCA can also be used in conjunction with an integral field spectrograph for imaging at visible wavelengths. OSCA provides a selection of 10 different occulting mask sizes from 0.25 - 2.0 arcsec and some with a novel gaussian profile. There is also a choice of 2 different Lyot stops (pupil plane masks). A dichroic placed before the AO system can give us improved nulling when occulting masks larger than the seeing disk are used. We also present results from initial testing and commissioning at the William Herschel Telescope.

The use of faint reference stars for the selection of good short exposure images has recently been demonstrated as a technique which can provide essentially diffraction-limited I band imaging from well-figured ground-based telescopes as large as 2.5 m diameter. The faint limiting magnitude and enhanced isoplanatic patch size for the selected exposures technique means that 20% of the night sky is within range of a suitable reference star for I-band imaging. Typically the 1%-10% of exposures with the highest Strehl ratios are selected. When these exposures are shifted and added together, field stars in the resulting images have Strehl ratios as high as 0.26 and FWHM as small as 90 milliarcseconds. Within the selected exposures the isoplanatic patch is found to be up to 50 arcseconds in diameter at 810 nm wavelength. Images within globular clusters and of multiple stars from the Nordic Optical Telescope using reference stars as faint as I~16 are presented. The technique relies on a new generation of CCDs which provide sub-electron readout noise at very fast readout rates. The performance of the selection technique for various astronomical programs is discussed in comparison with natural guide star Adaptive Optics (AO).

We present results of our survey of faint companions to O-stars using the adaptive optics (AO) system on the 3.63-meter Advanced Electro-Optical System (AEOS) telescope at the summit of Haleakala, on the island of Maui. The AEOS telescope is part of the United States Air Force's Maui Space Surveillance Site. We have surveyed most of the O-stars brighter than V magnitude 8.0 in the declination range of -25 to +65 degrees for faint companions. We are using the I-band (800 nm central wavelength, 150 nm approximate FWHM) for the survey. This is done for two reasons: 1) the distinctly red filter will de-emphasize the O-star primary and enhance the faint (presumably redder) secondary, increasing the dynamic range; and 2) using I-band allows all of the shorter wavelength light to be sent to the AO system, increasing its performance for fainter stars. We describe the scientific results of our survey as well as the reduction process we used to generate relative photometric results from a 12-bit frame transfer camera with no native ability to generate a bias frame.

Rethinking the efficient use of 4m-class telescopes in the dawning era of larger facilities is a timely but challenging debate. The extensive use of PUEO for imaging (and now spectroscopy) has kept CFHT at the forefront of scientific research with adaptive optics since its commissioning in 1996. Even though larger facilities are now starting to think about ways to implement high order AO systems, we believe the medium size of the CFHT and the excellent quality of the site on Mauna Kea is a perfect combination to reach the highest performances with a high order AO system. The fields of application of high order adaptive optics are exciting: They include extremely high contrast imaging and coronography in the near-infrared and diffraction-limited imaging in the optical, with the corresponding gain in angular resolution. In this paper we present a quick description of a few specific astrophysical problems that would benefit from an upgraded AO system at the Canada-France-Hawaii Telescope. More technical details about the upgrade of PUEO are presented by Lai et al. and Cuillandre et al. in these proceedings, see papers 4839-78 and 4839-31.

We present a note on low to medium resolution spectroscopy using adaptive optics (AO) system. A special focus is put on the problem of spectral slope variations. In principle a stellar image compensated by AO has a varying point spread function (PSF) strongly dependent on the observing wavelength. Even when the AO is working perfectly, the fraction of the energy in a finite size slit will change with the wavelength. The performance of AO correction is very sensitive to the observing conditions. Spectral slope variations directly connected to the wavelength dependency of the enclosed energy in the slit. Those features common and relatively harmless in conventional spectroscopy such as temporal variation in the seeing, brightness of the targets, imperfect slit peaking, atmospheric differential refraction, and fixed aperture size at spectral extraction, all introduce artificial continuum slopes. The degree of uncertainty in the spectral slope could be serious enough to interfere the observing goals in AO spectroscopy. A case for a spectroscopic observation for low mass stars is presented to demonstrate the problem. We found a steep continuum slope that is unrealistic for a low mass star. We undertook laboratory experiments with a calibration source in the AO system to test if the unrealistic continuum slope could be accounted for by the varying AO performance. In the experiments the "bluing" of the continuum slopes have been confirmed when the light source is dropping off of the slit or the wavefront reference source is faint. The effects are also qualitatively reproduced with calculations done by an AO simulation code.

The breakthrough of silicon immersion grating technology at Penn State has the ability to revolutionize high-resolution infrared spectroscopy when it is coupled with adaptive optics at large ground-based telescopes. Fabrication of high quality silicon grism and immersion gratings up to 2 inches in dimension, less than 1% integrated scattered light, and diffraction-limited performance becomes a routine process thanks to newly developed techniques. Silicon immersion gratings with etched dimensions of ~ 4 inches are being developed at Penn State. These immersion gratings will be able to provide a diffraction-limited spectral resolution of R = 300,000 at 2.2 micron, or 130,000 at 4.6 micron. Prototype silicon grisms have been successfully used in initial scientific observations at the Lick 3m telescope with adaptive optics. Complete K band spectra of a total of 6 T Tauri and Ae/Be stars and their close companions at a spectral resolution of R ~ 3000 were obtained. This resolving power was achieved by using a silicon echelle grism with a 5 mm pupil diameter in an IR camera. These results represent the first scientific observations conducted by the high-resolution silicon grisms, and demonstrate the extremely high dispersing power of silicon-based gratings. New discoveries from this high spatial and spectral resolution IR spectroscopy will be reported. The future of silicon-based grating applications in ground-based AO IR instruments is promising. Silicon immersion gratings will make very high-resolution spectroscopy (R > 100,000) feasible with compact instruments for implementation on large telescopes. Silicon grisms will offer an efficient way to implement low-cost medium to high resolution IR spectroscopy (R ~ 1000-50000) through the conversion of existing cameras into spectrometers by locating a grism in the instrument's pupil location.

We have developed an instrument, MEDI (Massive Exoplanet Differential Imager), that takes advantage of a unique method of starlight rejection, simultaneous differential imaging, in order to directly image massive planets around nearby stars. Using this technique we expect to achieve suppression of starlight to the photon-noise limit, which means that increased exposure time will translate into higher sensitivities. This is in contrast to past sequential and two-color simultaneous studies that reach a sensitivity floor due to speckle-noise limitations. MEDI is currently installed in ARIES, the infrared camera that will be commissioned at the newly refurbished 6.5 MMT in January 2003, with the world’s first adaptive secondary. This should allow us to take Nyquist sampled, diffraction-limited images in the near-IR. The adaptive secondary will also give us unprecedented throughput while minimizing the thermal background and providing a smooth PSF. Based on lab results, we expect to be able to detect objects 10⁶ times fainter than their primaries at 0.5” separations in 2 hours, limited only by photon noise. This suggests that we will be sensitive to objects with masses as small as ~5 M_Jupiter at separations of greater than ~5 AU for G2 V stars that are ~300 Myr old and within about 10 pc. Therefore, we will probe a unique search space compared with current radial velocity methods, which are so far restricted to close-in (<6 AU) orbits.

The predominant effect of the atmosphere on the incoming wavefront of an astronomical object is the introduction of a phase distortion, resulting in a speckle image at the ground-based telescope. Deconvolution from wavefront sensing is an imaging technique used to compensate for the degradation due to atmospheric turbulence, where the point spread function is estimated from the wavefront sensing data. An accurate calibration of the wavefront sensor is critical to estimating the point spread function and hence reconstructing the object. This paper investigates the calibration of a Shack-Hartmann wavefront sensor used for deconvolution from wavefront sensing.

We have designed and built an infrared camera using a Rockwell HAWAII MBE array sensitive from 1-5 microns. This camera is optimized for sensitive imaging in the 3-5 micron wavelength range, i.e. the L’ and M photometric bands. When used with the deformable secondary adaptive optics (AO) system on the 6.5m MMT telescope, the camera will be ideal for direct imaging surveys for extrasolar planets around young, nearby stars. Based on the models of Burrows et al (2001), we calculate that in a 2-hour background-limited integration with MMT AO we will be able to detect, in both M and L’ bands, a planet of 1 billion year (Gyr) age and 5 Jupiter masses (MJ) at a distance of 10 parsecs (pc). Our simulations of atmospheric speckle noise suggest that background limited M and L’ observations are possible at about 1.5 and 2.5 arcseconds, respectively, from a solar-type star at 10pc distance. The speckle limits move inward dramatically for fainter stars, and brighter planets or brown dwarfs can be seen even where the speckles overwhelm the background noise. The camera opens up a region of parameter space that is inaccessible to the radial velocity technique, and thus the two methods are highly complementary.

The California Extremely Large Telescope (CELT) project has recently completed a 12-month conceptual design phase that has investigated major technology challenges in a number of Observatory subsystems, including adaptive optics (AO). The goal of this effort was not to adopt one or more specific AO architectures. Rather, it was to investigate the feasibility of adaptive optics correction of a 30-meter diameter telescope and to suggest realistic cost ceilings for various adaptive optics capabilities. We present here the key design issues uncovered during conceptual design and present two non-exclusive "baseline" adaptive optics concepts that are expected to be further developed during the following preliminary design phase. Further analysis, detailed engineering trade studies, and certain laboratory and telescope experiments must be performed, and key component technology prototypes demonstrated, prior to adopting one or more adaptive optics systems architectures for realization.

Rayleigh laser guide star technology is discussed here with particular attention paid to the effects of laser pulse length, a parameter that becomes more significant to the design when telescope apertures are greater than 10 meters. After reviewing the relative return signal for Rayleigh versus sodium laser guide stars, a brief review of the pulse length characteristics of sodium lasers is given. Only one of the proposed sodium laser systems is pulsed while the others are CW. To insure star-like sources at the wavefront sensor with FWHM < 1.0 arcsec, lasers that will be most useful for Extremely Large Telescopes must have a short pulse format whereas CW lasers will be of little to no use. A relatively simple Rayleigh laser guide star method is described for Ground Layer Adaptive Optics (GLAO). This method provides a way to average out the effects of high altitude turbulence with a single Rayleigh laser guide star leaves intact the wavefront sign needed to correct ground-layer wavefront perturbations.

Measurements of anisoplanatism from data obtained with natural guide star adaptive optics on the Lick Observatory 3m are presented. These were obtained from short exposures of binary stars with the IRCAL camera whose field of view (~20”) is generally considered isoplanatic in the K-band. However, measurable amounts of high-order anisoplanatism were present at separations of ~7” and ~12” with an isoplanatic patch size estimated to be ~26”. Within this field, there was measureable differential image motion between the binary star components. This image motion was small compared to the size of the diffraction-spot and therefore had negligible effect.

We have developed a system for real-time monitoring of the optical turbulence altitude profile in the atmosphere above large telescopes. A high order Shack-Hartmann wavefront sensor is used to observe bright binary stars, and the turbulence profile is recovered from a cross-correlation of the measured wavefront slopes. This method is referred to as SLODAR (SLOpe Detection And Ranging), by analogy with the well-known SCIDAR binary star scintillation profiling technique. SLODAR can be implemented using relatively simple and inexpensive hardware. The monitor described here is based on a commercially available unintensified Firewire digital video camera. The system has been installed at the 4.2m William Herschel telescope in La Palma and will be used to support observations with adaptive optical correction at the telescope. The turbulence data will allow optimisation of range-gate altitudes for Rayleigh laser beacons, deformable mirror conjugation altitudes and other AO system parameters, as well as characterisation of the AO corrected point spread function for anisoplanatism. The system will also provide statistical site characterisation data to permit accurate modeling of future AO instruments.

Limiting magnitude of A.O. reference stars is set by wavefront sensor intrinsic noise. Recently available avalanche intensified CCD detectors allow single photon event detection (photon counting) virtually free from readout noise. The paper describes a program started in spring 2002, aiming to use a Marconi L3CCD as a wavefront sensor for A.O.

The recently commissioned GroundWinds LIDAR Observatory, based at ~3300 m on the slope of Mauna Loa, can measure altitude resolved line-of-sight wind velocities, turbulence power spectra, aerosol content and faint cirrus clouds among other things of interest to astronomers. The overarching goal of the GroundWinds program is to develop and demonstrate incoherent ultra-violet LIDAR technology for a future space-based system to measure the vertical structure of global winds from molecular backscatter. The LIDAR observatory employs spectral line profiling of incoherent backscattered 355 nm laser light. Rapid measurement of the Doppler shift (400 ns resolution) is accomplished by feeding the returned laser light into a combination of two Fabry-Perot etalons and collapsing the interference fringes into a 1-dimensional interference pattern using a conical optic. This allows the system to obtain the maximum signal-to-noise ratio and best vertical resolution given the performance of the CCD. Each measurement takes 10 s. The molecular return is strong up to 15-km altitude. The YAG laser is pulsed at 10 Hz, and each pulse is stretched to 50 ns; the average power dissipated is 5 W. The outgoing beam is expanded to match the field of view of the telescope. The Doppler shift as a function of altitude, measured along two lines of sight orthogonal to one another, is then used to determine the horizontal wind velocity as a function of altitude. A recent intercomparison campaign demonstrated the accuracy of the GroundWinds instrument. In addition to average wind measurements intended for global winds, the LIDAR can be operated with a short integration time and used to directly measure turbulence spectra over a range of elevations. The turbulence spectra are used to approximate the velocity turbulence parameter, C_v², and turbulent dissipation. A recent comparison with an independent measurement of C_T² has shown good agreement. Data from the incoherent LIDAR are used in a custom forecasting project (Mauna Kea Weather Center: http://hokukea.soest.hawaii.edu) that provides operational support for the world-class group of astronomical observatories located on the summit of Mauna Kea. The LIDAR data are used to help prepare wind and turbulence nowcasts/forecasts for the summit of Mauna Kea (~4000 m) and as input for an operational mesoscale numerical weather prediction model (MM5). Clear-air turbulence in both the free atmosphere and in the summit boundary layer causes phase distortions to incoming electromagnetic wave fronts, resulting in motion, intensity fluctuations (scintillation), and blurring of images obtained by ground-based telescopes. Astronomical parameters that quantify these effects are generically referred to as seeing. Seeing improves or degrades with changes in the vertical location and strength of turbulence as quantified by profiles of the refractive index structure function C_n². C_n² fluctuations usually occur at scales that are too small for routine direct measurement, but they can be parameterized from vertical gradients in wind, temperature, and moisture in our MM5 runs. Seeing at a particular wavelength is then calculated by vertically integrating the C_n² profile. LIDAR wind profiles represent an important data resource for nowcasting seeing, input for MM5 initial conditions and algorithm refinement, and for forecast verification.

We present here results using two novel active optic elements, an electro-static membrane mirror, and a dual frequency nematic liquid crystal. These devices have the advantage of low cost, low power consumption, and compact size. Possible applications of the devices are astronomical adaptive optics, laser beam control, laser cavity mode control, and real time holography. Field experiments were performed on the Air Force Research Laboratory 3.6 meter telescope on Maui, Hawaii.

This presentation reports the numerical simulations we have done in order to evaluate the performance of the first-light AO system of LBT. The simulation tool used for this purpose is the Software Package CAOS, applicable for a wide range of AO systems and for which a brief recall of the main features is made. The whole process of atmospheric propagation of light, wavefront sensing (using a complete model of the pyramid wavefront sensor), wavefront reconstruction using the LBT672 adaptive secondary mirror modes), and closing of the loop, is simulated. The results are given in terms of obtained Strehl ratios in J-, H-, and K-band. Estimation of the resulting sky-coverage in K-band for different regions of the sky are also expressed. A comparison with the performance that would be obtained by using a Shack-Hartmann sensor is presented, confirming the gain achievable with the pyramid sensor.