Active and adaptive optics technology has emerged from the laboratory and is being applied to improve the performance of optical imaging and laser systems. In the last few years, development of both systems and components has accelerated. Many new concepts and devices have appeared, among which are high-performance deformable mirrors, new types of wavefront sensors, and more sophisticated wavefront processing algorithms. Equally important, a better understanding of the system design aspects of adaptive optics has been reached, particularly of the need for optimizing each system according to its application. For example, the dominant requirement in laser systems is to achieve a high Strehl ratio, whereas for ground-based astronomy the availability of guide stars is a major concern. Current developments in adaptive optics for ground-based astronomy include the use of IR wavelengths, partial wavefront compensation using natural guide stars, and the use of laser guide stars to allow all-sky coverage with full compensation at visible wavelengths. While progress to date has been impressive, much work remains to bring this technology into general use.

Modeling of deformable mirror (DM) surfaces is reviewed and two different types of DM models are compared. The first model creates the DM surface through the superposition of an influence function over the extended actuators. The second model uses a two-dimensional cubic spline fit over the extended actuators to create the DM surface. The ability of the predicted DM surface to correct for different spatial frequencies was investigated. It is shown that correctability may vary in both magnitude and shape over the applicable range of frequencies. Results indicate that the bi-cubic spline model is the easiest and fastest to use and it may be extended to incorporate more advanced features.

The optical transfer function (OTF) is examined for an optical instrument corrected by an adaptive optics systems. A model introducing the application of modal control to the OTF and accounting for the effects of time lags and angular depointings is presented. The model allows the estimation of OTF decay in respect to the Fried diameter and for different time lags in an adaptive mirror. Modal control is described and delays in the controls of the adaptive mirror system are discussed.

An adaptive optical system dedicated to high resolution imaging can be modelized in terms of transfer loops. This model permits one to estimate the response of such a system to time-varying wavefront perturbation. A block diagram describing the closed loop of the adaptive optics of the Come-On project is given and a theoretical expression expressed in terms of the Z transform is found for the case of a nodal approach. In a second step, identification methods are used to determine the best parameter values of these nodal models. Finally, the responses of these models to known perturbations are compared with the experimental data recorded during the Come-On experiments. The same analyses are conducted for modal models. In these two cases, the results are found to be in good agreement.

A technique for extracting the relative phase (piston) differences between telescopes of a phased array or segments of a mirror is examined. The analysis concentrates on examination of measuring the relative phase of three segments with a single far-field intensity distribution. The contour of the first minimum in the far-field diffraction spot is directly related to the relative phase differences between three segments or telescopes. The paper will address various methods for extracting the information and using it to phase the segments or telescopes. Restrictions to the technique and questions regarding uniqueness of the solution are addressed. Limitations due to finite resolution in the far-field, segment or telescope tilt, dynamic beam jitter, and higher-order aberrations are quantified by computer simulation. Methods of extending the technique to real-time control are proposed.

Current adaptive optical telescope designs use a single deformable mirror (DM), usually conjugated to the aperture plane, to compensate for the cumulative effects of optical turbulence. The corrected field of view of an adaptive optics system could theoretically be increased through the use of multiple DMs conjugated to a number of corresponding planes which sample the turbulence region in altitude. Control of each DM is via a method for determining the phase distortion contributed by atmospheric layers at the selected altitudes. A theoretical analysis for determining these phase contributions takes advantage of the spatial diversity of wavefront sensor measurements from two or more reference sources. These separate wavefront sensor measurements are processed via minimum mean square error filtering to yield an estimate of the phase perturbation caused by a particular turbulent layer of the atmosphere. Our initial investigation indicates that multiple wavefront corrector adaptive optics systems will require much brighter reference sources than single wavefront corrector systems.

The use of laser guide stars in conjunction with adaptive optical telescopes offers the possibility of nearly diffraction limited imaging performance from large, ground-based telescopes. In this paper we investigate the expected imaging performance of an adaptive telescope using laser guide stars created in the mesospheric sodium (Na) layer. A two to three meter class telescope is analyzed for the case of a single, on axis guide star at an altitude of 92 km (nominal height of the mesospheric Na layer). We assume the telescope pupil is annular with approximately 15 wave front sensor subapertures and mirror actuators spanning the pupil diameter. Imaging performance is quantified in terms of the pupil averaged rms wave front error, the optical transfer function, the point spread function, the Strehl ratio and finally the angular resolution. The performance analysis takes into account the degradation caused by the limitation of the wave front sensor as well as the deformable mirror. These limitations include the finite spacing and size of the wave front sensor subapertures and the spacing and influence function of the mirror actuators. The effects of anisoplanatism and shot noise are also included in the analysis. The results of the investigation indicate that a 3 meter adaptive telescope using a single Na guide star is capable of achieving a Strehl ratio of 0.57 and an angular resolution nearly matching that of diffraction limited performance (0.05 arcsec). This performance is achieved assuming ro equals 20 cm and a 5 watt laser is used to create the guide star. The effect of variations in seeing conditions and guide star brightness are also investigated.

The effects of focal and angular anisoplanatism are computed in order to evaluate the utility of using a single laser-produced guide star for the correction of ground-based astronomical imaging. The equations for the calculation of these effects are derived and the performance of a sodium-layer laser guide star system is computed for a Hufnagel atmospheric turbulence model. These results are presented in scaled units and for selected telescope apertures from 2-8 m in diameter operating at wavelengths from 0.5 to 2.0 microns. The limiting telescope aperture size which can be adequately corrected using a single sodium-layer guide star is shown to be much larger than previously estimated.

Experiments have been conducted in the atmosphere above Mt. Laguna Observatory to measure the properties of laser guide stars. The experimental system consists of a high frame rate video camera which records the backscattered light from an Excimer laser working in the near-UV at 351 rim. The Mt. Laguna 1-meter telescope is used to both transmit the outgoing beam and to image the return beam. The outgoing laser pulse triggers a time-gated image intensifier within the video camera which, with an appropriately selected time delay, records a time slice of the backscattered return signal. Preliminary results from the experiment can be used to calibrate the laser power needed to operate a large ground-based adaptive optics telescope.

We describe here numerical methods for reducing data taken with curvature sensing measurements. The Laplacian integration, spherical aberration estimate and removal, automatization of image parameter determination routines are explained. We then described the user instrument to do real time telescope testing.

Deformable mirror is the key element for adaptive optical wavefront correction. The number of actuators decides the complexity and cost of adaptive optical system. In this paper computer simulations of wavefront error for fitting different Zernike terms by deformable mirror with different number of actuators are presented. The arrangement of actuator and the influence function of mirror are discussed in respect of fitting error. The minimum number of actuators for fitting different Zernike orders of wavefront are given. Some optical experiments of fitting capability have been done with 19 and 37-element deformable mirrors and a Zygo interferometer.

An adaptive optics scheme is proposed which uses curvature sensing, a bimorph PZT deformable mirror, a simple extrafocal image splitter and a sensitive, low noise Solid State Photomultiplier array detector. With curvature sensing, the incoming wavefront is described in terms of curvature which can be directly mapped point-by-point to a driven deformable segmented mirror. While other methods such as Hartmann, require time and cost consuming computations, the curvature mirror can be driven in near real-time taking full advantage of the 11 kHz bandwidth of the bimorph mirror. The SSPM is easily configured to any array pattern without loss of performance. The arrangement is particularly suited to large telescopes in the 8 m class for which the wavefront curvature sensor can provide additional useful information to correct for collimation and structural variations as well as the rapid seeing correction.

The MMT consists of six comounted 1.8 m telescopes from which the light is brought to a combined coherent focus. Atmospheric turbulence spoils the MMT diffraction-limited beam profile, which would otherwise have a central peak of 0.06 arcsec FWHM, at 2 microns wavelength. At this wavelength, the adaptive correction of the tilt and path difference of each telescope beam is sufficient to recover diffraction-limited angular resolution. Computer simulations have shown that these tilts and pistons can be derived by an artificial neural network, given only a simultaneous pair of in-focus and out-of-focus images of a reference star formed at the combined focus of all the array elements. We describe such an adaptive optics system for the MMT, as well as some successful tests of neural network wavefront sensing on images, and initial real-time tests of the adaptive system at the telescope; attention is given to a demonstration of the adaptive stabilization of the mean phase errors between two mirrors which resulted in stable fringes with 0.1 arcsec resolution.

A 19-segment adaptive mirror system for use in solar astronomy has been developed and operated on the Sacramento Peak Tower Telescope. The system has proven itself to be capable of improving the quality of an image, at times achieving 1/3 arcsecond resolution in 1-3 arcsecond seeing conditions.

The Johns Hopkins University is developing a stellar coronagraph which will use adaptive optics to achieve nearly diffraction-limited imaging at optical wavelengths with 1-2 meter class telescopes. The first phase of development, the incorporation of an image motion compensation system into the coronagraph, is complete. Performance tests have resulted in a factor of 2 gain in image resolution, corresponding to the maximum gain predicted by theory. The next phase of development involves the construction of an electrostatically deformable membrane mirror and a wavefront curvature sensor for the removal of higher order aberrations. A membrane mirror with 91 actuators has been built for laboratory testing. Integration of the adaptive mirrors, high speed wavefront sensor, and control processor is forthcoming.

An active optics system for a 3.5-meter f/1.75 borosilicate honeycomb mirror has been designed and built. The system hardware and software are described, and preliminary test results are presented that demonstrate the structured mirror responds well to the active optics control. Plans for extensive further testing are described. The results of the testing will guide a redesign of the system, before installation of the second-generation system in the WIYN Telescope, to be built on Kitt Peak in Arizona.

We present a tracking system that stabilizes atmospheric and instrumental image motion in vacuum tower telescopes. The system is designed to lock on low contrast features, such as solar granulation or other small scale structure. A matrix diode array rapidly scans the scene of interest, usually with a field of 5 arcsec. Images are cross-correlated in real time with a previously recorded reference image of the same area. The drive signal for the image motion corrector, a small, articulated mirror, is generated by measuring the position of the cross correlation maximum. Reference pictures are updated every 30 s in order to adapt to the changing small scale solar features. Performance tests show that the residual image motion in the tracked image is 0.05 arcsec rms compared to a typical 0.5 arcsec rms for the untracked image. The system locks on any small scale structure anywhere on the sun. The bandwidth of the servo system is 40 Hz, or sufficient to stabilize image motion on a meter-class solar telescope.

Two alternative methods have traditionally been used to maintain telescope alignment and focus. The direct method measures rigid body changes (decentration, tilt and despace) using internal sensors and predicts the resulting performance degradation. The indirect method measures the performance degradation with a wavefront sensor and predicts the rigid body changes. An alternative indirect method uses an image sharpening merit function value derived from the image irradiance data . The image sharpening function has the property that it achieves its maximum value when the image has no residual aberrations. An aberrated image can be improved by introducing aberration corrections using an active mirror until the merit function value is optimized. Therefore, by forming a closed loop between the image detector and the active optical element, a telescope can be realigned and refocused to optimum performance without the use of a separate wavefront sensor or internal alignment/focus sensors. A comparison of the direct and indirect methods and recent experimental results of the image sharpening method are presented.

The long baselines of the next-generation ground-based optical stellar interferometers require optical delay lines which can maintain nm-level path-length accuracy while moving at high speeds. NASA-JPL is currently designing delay lines to meet these requirements. The design is an enhanced version of the Mark III delay line, with the following key features: hardened, large diameter wheels, rather than recirculating ball bearings, to reduce mechanical noise; a friction-drive cart which bears the cable-dragging forces, and drives the optics cart through a force connection only; a balanced PZT assembly to enable high-bandwidth path-length control; and a precision aligned flexural suspension for the optics assembly to minimize bearing noise feedthrough. The delay line is fully programmable in position and velocity, and the system is controlled with four cascaded software feedback loops. Preliminary performance is a jitter in any 5 ms window of less than 10 nm rms for delay rates of up to 28 mm/s; total jitter is less than 10 nm rms for delay rates up to 20 mm/s.

Segmented reflectors have been proposed for space-based applications such as optical communication and large-diameter telescopes. An actuation system for mirrors in a space-based segmented mirror array has been developed as part of the National Aeronautics and Space Administration-sponsored Precision Segmented Reflector program. The actuation system, called the Articulated Panel Module (APM), articulates a mirror panel in 3 degrees of freedom in the submicron regime, isolates the panel from structural motion, and simplifies space assembly of the mirrors to the reflector backup truss. A breadboard of the APM has been built and is described. Three-axis modeling, analysis, and testing of the breadboard is discussed.

The Italian GALILEO telescope (TNG) is in an advanced phase of construction. Among the various new technical aspects of this telescope the active optics system is now receiving special consideration. In particular, the optical and informatic groups are considering the definition of the control environment dedicated to the active-optics. A solution based on an array of interconnected l6bit transputers is here described with the main requirements for the inter-communication and monitoring software.

High precision movement of telescope mirrors in two-mirror telescopes is examined. Active optics are used for high precision movement of M2 during exposures to introduce a specific amount of decentering coma. It is shown that the decentering coma may be controlled by high accuracy rotation of the mirror around a point near the center of curvature of M2. Basic equations are developed for the displacement of the barycenter of the geometric blur. A relation is obtained for determining the correct position around which the rotation should be performed. The application of these techniques to two-mirror telescopes is discussed. Ray tracing and analytical results were compared and showed good agreement.

The adaptive optics system being developed to sharpen images produced by telescopes at Mauna Kea is discussed. An approach based on new components developed and optimized for astronomical applications is described. The approach is limited to low-order wavefront compensation and is used for image stabilization. Avalanche photodiodes were used as sensors and reference stars were employed for sensing wavefront errors in a novel sensing technique based on wavefront curvature measurements. The instrument is described and expected performance is discussed.

A review of computer simulation of a low-order adaptive system based on curvature sensing and a bimorph mirror is given. The system was designed to accurately follow the lowest order Karhunen Loeve modes of the atmosphere. The purpose of the simulation was system design improvement and the determination of system performance limits. The four main areas of the simulation are described; atmospheric simulation, Fresnel propagation, bimorph model, and sensor feedback model. Results are presented and discussed.

The structure of a wavefront sensor suitable for adaptive optics is described. The sensor scans rapidly between extra focus images with a vibrating membrane mirror acting as a variable curvature active optical element. The membrane mirror and driver are described. The optical system is coupled to an array detector optimized to match the response of a bimorph mirror and directly produces wavefront error signals. Results are presented and compared with other techniques for curvature sensing. Tip-tilt corrections achieved using the wavefront sensor are discussed.

A telescope aperture of 2.2-m on Mauna Kea that routinely experiences d/r sub 0 = 4 in the near-IR can achieve a factor of 2 gain in angular resolution by tip-tilt correction of atmospheric-induced wavefront errors. To utilize the gains possible from tip-tilt correction, collimation errors and focus errors must also be removed. For its 2.2-m f/31 telescope, the University of Hawaii is in the process of implementing a five-axis fast guiding secondary consisting of a fast steering mirror platform and slow remote detilt, decenter, and despace collimation and focus drives. The near-term goal is to implement closed-loop tip-tilt image motion correction with open-loop collimation and focus control. The long-term goal is to add closed-loop collimation and focus control. This paper documents the progress to date on the fast steering mirror platform and its spider support structure.

Adaptive optics is one of the main features of the Very Large Telescope (VLT) of the European Southern Observatory (ESO) - an array of four 8 meter telescopes. These telescopes can be operated individually, in an incoherent and in a coherent interferometric beam combination mode. Each telescope will be equipped with adaptive optics systems for real-time correction of atmospheric turbulence effects. First results with a prototype system developed for the VLT demonstrated the feasibility and the significant gain of this technology for astronomical imaging. This paper describes the VLT adaptive optics system and its implementation program.

The latest developments of active optics of the ESO NTT include the reduction of friction in the lateral supports of the primary mirror and in the positioning system of the secondary mirror. The most important remaining problem is the local air condition. The implications for the ESO VLT and the latest developments in the design of its active optics are discussed.

This paper is a presentation of the Come-On-Plus adaptive optics system, based on the Come-On prototype. Come-On-PIus will be set up in 1992 on the ESO 3.6 m telescope in La Silla (Chile). It is an upgrade of the Come-On instrument, with a 52 actuator deformable mirror, and 30 Hz correction bandwidth. But the main improvement concerns the wavefront sensing, designed in this instrument for astronomical applications, with a high detectivity wavefront sensor and a specific mirror control algorithm. This system is planned for routine astronomical observing as well as providing design parameters for the adaptive optics system of the ESO Very Large Telescope (VLT).

This paper reports the results of the observations made with the VLT Adaptive Optics Prototype System 'COME-ON' at the ESO 3.6 meter telescope. The analysis of uncorrected and corrected images in the near-IR wavelength range (below 5 microns) leads to a detailed assessment of the system performance in terms of improvement of angular resolution that nearly achieves the ideal diffraction profiles down to 1.7 micron wavelength, as well as a Strehl ratio approaching 0.6-0.8 at 3.8 microns. A resolution of 0.12 arcsec has been obtained with this system at 1.7 microns which is wavelength-dependent on the temporal parameters of the observation. The current limiting magnitude for the reference source is m sub R = 11.5 when applying the full correction capabilities of the system, and m sub R = 13 if only the wavefront tilt is corrected.

The MARTINI adaptive optics system has been engineered for regular astronomical observations on the 4.2 m William Herschel Telescope on La Palma. The design specifications for such a general purpose image-sharpening device are discussed. Examples of the imaging performance achieved during development of MARTINI are presented, together with results from visible-region astronomical imaging programs. The prospects for spectroscopic observations with this class of system are outlined.

The potential of active structures technology for future IR and sub-mm telescopes deployed in space is discussed. A space infrared telescope with an aperture of at least 8 m, operating at about 50 k with surface precision better that 0.5 microns is required for future work in space IR and sub-mm astronomy. Telescope parameters, noise, and sensitivity are discussed. The effects of large scale figure errors are examined and conceptual designs of future space IR and sub-mm telescopes are presented. Possible uses of active structures technology and feedback control in future IR and sub-mm space observatories are suggested.

This paper describes experimental research being performed at the Jet Propulsion Laboratory to develop and validate control concepts arising out of NASA's Control Structure Interaction program. The facility is meant to be a ground testbed with relevance to a broad class of future precision optical space systems. The objective of the experimental program is to investigate a multi-layer control approach to the maintenance of nanometer level optical pathlength control in the presence of external disturbances and multiple structural resonances. A brief overview of the experimental facility is presented. The control design methodology is discussed, and several experimental results are presented.

A class of proposed space-based astronomical missions requiring large baselines and precision alignment can benefit from the application of Controlled Structures Technology. One candidate mission, that of a 35 meter baseline orbiting optical interferometer, is studied as a focus mission for a testbed for controlled structures research. Interferometry science requirements are investigated and used to design a laboratory testbed which captures the essential architecture, physics and performance requirements of a full scale instrument. Testbed hardware used for identification and control is presented, including an on-board six-axis laser metrology system using state of the art cat's eye retroreflectors. The testbed and research program are discussed in terms of controlled structures design and in terms of the expected benefits to the optical engineering and science communities.

The goal is to develop and demonstrate the potential benefits of applying the controlled structures technology (CST) approach to active and adaptive optics. The potential advantages were outlined in a previous paper. Two testbeds are envisioned for the experimental work. CST techniques are demonstrated experimentally on the first, a simple deformable mirror testbed (DMT) incorporating piezoelectric sensors and actuators and a single optical displacement sensor. Implementation of passive damping augmentation, local control and global high authority control is demonstrated. Some of the key issues in the approach to the control of larger interferometers or lightweight Cassegrain telescopes are discussed. Finally, some of the open areas for future research are summarized.

An overview of three ongoing TRW projects in the area of active structures is presented. The first project involves the development and validation of active member technology for future use in space based optical systems. The autonomous identification of changes in active structures and updating local compensator parameters for maximum performance is investigated in the second project. The third project is concerned with system level applications of active members to slewing and maneuvering aircraft. Mechanical, structural dynamics, and electronics tests of active members are described. Systems identification, compensation design, and performance simulation in active structure monitoring is reviewed and precision control of agile spacecraft is discussed.

A concept for correcting long wave low order distortions of a lightweight composite mirror for space applications is described. One of the attractive features of this concept for space applications is that a backup structure is not required. The actuation system consists of piezoelectric elements, attached directly to the back of the mirror surface. The system is self balancing. This paper describes the test results for a one-half meter curved composite reflector.

Wavefront control modeling is an important technique for examining the steady state performance and design tradeoffs of many types of beam control systems. The present paper discusses a model for simulating a wide field of view laser projection beam control brassboard experiment. The simulation is designed to optimize system performance over the full field of view during brassboard operation. Simulation results will be utilized to define a minimum set of standard viewing angles at which the brassboard experiment will need to be calibrated. During the experiment, the computer software will interpolate between these discrete calibration angles to obtain the information required to operate at any angle over the full field of view. In the brassboard experiment wavefront sampling is achieved with a Hartmann type outgoing wavefront sensor, which receives slope information from 48 holographic optical elements on the primary minor. Optical aberrations are corrected by a 97 actuator deformable mirror. Key elements of the simulation are: 1) a deformable minor model constructed from NASTRAN generated actuator influence functions, 2) an outgoing wavefront sensor model which computes average slope values over the holographic optical elements, and 3) submodels for all budgeted beam control subsystem, sensor and diagnostics error sources. Critical analyses discussed include the characterization of the calibration/interpolation geometry and the expected performance effects as a function of field of view.

Many side-pumped lasers exhibit significant index gradients across the gain region aperture. For pulsed lasers where these gradients are time dependent, extraction with good beam quality requires the use of an adaptive optic. Since these inhomogeneities are systematic, wavefront correction can be performed with a model deformable mirror. We have designed a resonator which uses a cylindrically deformable mirror to correct for wavefront aberrations in a pulsed nuclear-reactor-driven laser. The mirror is capable of correcting up to ten waves of cylindrical focus error while maintaining tip/tilt alignment of the resonator. It is based around the flat plate bending using magnetostrictive actuators. A cylindrical intracavity beam expander is used to image the DM into the laser gain region. The beam expander can be adjusted to vary the resonator magnification in one axis, or to control the stability of the resonator. The mirror is controlled closed loop using a four channel wavefront sensor and a digital control system.

We have developed an adaptive optics system that corrects up to five waves of 2nd-order and 3M-order aberrations in a high-power laser beam to less than 1/10th wave RMS. The wavefront sensor is a Hartmann sensor with discrete lenses and position-sensitive photodiodes; the deformable mirror uses piezoelectric actuators with feedback from strain gauges bonded to the stacks. The controller hardware uses a VMIE bus. The system removes thermally induced aberrations generated in the master-oscillator-power-amplifier chains of a dye laser, as well as aberrations generated in beam combiners and vacuum isolation windows for average output powers exceeding 1 kW. The system bandwidth is 1 Hz, but higher bandwidths are easily attainable.

Results from a benchtop experiment to demonstrate phase compensation using a 512 segment, 1500 degree-of-freedom adaptive optic system are presented. Atmospheric phase distortion is simulated by a static Kolmogorov spectrum aberration plate with r0 equal to the subaperture size. The phase gradients are measured using a Poisson-limited, self-referenced shearing interferometer which operated at two distinct shear lengths. A parallel processor is then employed utilizing a sparse matrix multiply to reconstruct the phase front in realtime. The performance of the compensation was determined by measuring the normalized half lambda/D intensity ratio in the Fourier transform plane. Corrections to a Strehl ratio of 0.55 were performed, consistent with the measured sensitivity of the system.

Predetection compensation combined with post detection image processing for the case of imaging through atmospheric turbulence is addressed. Full and partial predetection compensation using adaptive optics is combined with bispectrum speckle imaging post-processing, and performance improvements are assessed. Full compensation was found to provide a large improvement in the signalto- noise ratio (SNR) of the power spectrum estimate compared to the uncompensated case. Lower degrees of correction provided smaller improvements in the power spectrum SNR, and a very low degree of compensation provided results indistinguishable from the uncompensated case. Three regions of performance improvement were found with respect to the object Fourier phase spectrum estimate: 1) the fully compensated case, where bispectrum post processing provided no improvement in the phase estimate over that obtained from a fully compensated long exposure image; 2) a partially compensated regime, where applying bispectrum post processing to the compensated images provided phase spectrum estimation superior to the uncompensated bispectrum case; and 3) a very poorly compensated regime, where the results were essentially indistinguishable from the uncompensated case. Previously validated simulation codes were used to conduct this investigation.

We are incorporating a novel self-referencing Mach-Zehnder interferometer into a large scale laser system as a real time, interactive diagnostic tool for wavefront measurement. The instrument is capable of absolute wavefront measurements accurate to better than XIlOpv over a wavelength range > 300 nm without readjustment of the optical components. This performance is achieved through the design of both refractive optics and a catadioptric collimator to achromatize the Mach-Zehnder reference arm. Other features include polarization insensitivity through the use of low angles of incidence on all beamsplitters as well as an equal path length configuration that allows measurement of either broad-band or closely spaced laser-line sources. Instrument accuracy is periodically monitored in place by means of a thermally and mechanically stable wavefront reference source that is calibrated off-line with a phase conjugate interferometer. Video interferograms are analyzed using Fourier transform techniques on a computer that includes a dedicated array processor. Computer and video networks maintain distributed interferometers under the control of a single analysis computer with multiple user access.

In this paper an extension of the principle of the Hartmann-Shack wavefront sensor to the neasureirtent of the sphericity of a specular reflection surface is presented. The features of this widely usable system are discussed by means of the application of this approach to the modeling of surface aberrations of the cornea of the eye. The results of measurements performed with a first prototype on a set of stainless steel calibration spheres of different radii are shown.

JPL's PSR program is developing enabling technologies for large space telescopes employing segmented optics, and in particular, lightweight, thermally stable mirrors for telescopes operating at sub-mm wavelengths. JPL's vacuum cryointerferometric optical test facility includes a computer-controlled phase shifting interferometer operating at 10.59 microns and laser metrology for measurement of piston and absolute radius of curvature. The optical metrology has been integrated with a large vacuum chamber including a liquid nitrogen shroud capable of radiatively cooling 1.5 meter aperture mirrors down to 150 K. The facility is now being used to characterize prototype mirrors in both their room-temperature 'as-replicated' condition, and under orbital temperature conditions. The performance and limitations of the optical metrology hardware and software components are noted. Representative test results on prototype one meter-class composite mirrors being developed for PSR and related ground-based programs are also presented.

Space-based interferometers and large segmented reflectors will require precision metrology with nanometer accuracy for active control systems. This paper summarizes an experimental study of an optical heterodyne interferometer with resolution of about 3 nm. Optical signals are carried to and from the interferometer by optical fibers to avoid thermally loading the reflector panels. Heterodyne detection provides far greater precision than does traditional interferometry.

Foucault knife edge testing has been the classic qualitative test method for optical surface and wavefront quality measurement for over one hundred years. We have developed an innovative method, using off the shelf CIigit.a1 microprocessors and a solid state camera, whereby quantitative evaluation of surfaces has been achieved. This paper will describe a method that upgrades the knife edge test from a qualitative test to a precise quantitative evaluation. The hardware is described and results are shown comparing the knife edge test with interferometric measurements.

This paper presents the measurement of the wave aberrations of human eyes with a Hartmann-Shack Sensor (HSS), used in a system for active optical depth resolution improvement of the Laser Tomographic Scanner(LTS). A least-squares algorithm of modal wavefront estimation from the tested derivatives is described. The noise propagation of this algorithm is examined. And the experimental results of tested living eyes are presented.