An astrometric interferometer in space will provide a dramatic improvement in the accuracy with which stellar positions may be measured. The potential and the far ranging impact in science of this capability has deeply impressed the scientific community, with the result that a high priority has been assigned to a mission for precision astrometry. In this talk I will review the origins of the AIM concept, describe the scientific opportunities, and show the relationship to ground capabilities and to other space astronomy missions.

POINTS, an astrometric optical interferometer with a nominal measurement accuracy of 5 microarcseconds for the angle between a pair of stars separated by about 90 deg. is presently under consideration by two divisions of NASA-OSS. Based on a preliminary indication of the observational needs of the two missions, we find that a single POINTS mission will meet the science objectives of both TOPS-1 and AIM. The instrument detects a dispersed fringe (channelled spectrum) and therefore can tolerate large pointing errors. In operation, the difficult problem of measuring the angular separation of widely spaced star pairs is reduced to two less difficult problems: that of measuring the angle between the two interferometers and that of measuring interferometrically the small offset of each star from the corresponding interferometer axis. The question of systematic error is the central theme of the instrument architecture and the data-analysis methods. Stable materials, precise thermal control, and continuous precise metrology are fundamental to the design of the instrument. A preliminary version of the required picometer laser metrology has been demonstrated in the laboratory. Post-measurement detection and correction of time-dependent bias are the essential elements in data analysis. In that post-measurement analysis, individual star-pair separations are combined to determine both the relative positions of all observed stars and several instrument parameters including overall time-dependent measurement bias. The resulting stellar separation estimates are both global and bias-free at the level of the uncertainty in the reduced (i.e., combined and analyzed) measurements.

The purpose of this research is to explore, from the point of view of nonlinear filtering, the feasibility of microarcsecond astrometry using space-based optical interferometers in typical disturbance environments. Two nonlinear estimators are developed to enable high precision estimates of the optical path difference (OPD) between the two starlight paths in a spaceborne optical interferometer. Focal plane fringe data measurements by either CCD (charge coupled device) or photon counting (PC) cells are processed by an Extended Kalman Filter (EKF)- based algorithm to yield the OPD estimate. Whereas the filter based on CCD cell measurements results in a straightforward implementation of an EKF, the filter utilizing PC cells requires some innovations before yielding to the EKF framework. Instead of using the photon arrival events as measurements, the filter measurements are taken to be the interarrival time between photons. The excellent results obtained with the PC filter are backed by intuition based on linear analysis results. Simulation studies show that as the CCD readout time is decreased, the performance of the CCD filter approaches that of the PC filter.

The Precision Optical Interferometer in Space (POINTS) is a free flying, space-based, astrometric interferometry mission employing an instrument with two 2-meter baseline interferometers. This paper addresses a disturbance analysis of the preliminary POINTS spacecraft and instrument design in order to determine whether and to what extent any isolation and/or added structural damping is necessary to meet the POINTS instrument requirements. The analysis was performed using the Integrated Modeling of Optical Systems (IMOS) integrated modeling tool. IMOS is an integrated software environment wherein structural, optical, and control system modeling can be performed. Linearized optical models, structural finite element models, and disturbance characterization models were developed and integrated in IMOS. Starlight fringe coherence was used as a metric to quantify the performance of the POINTS instrument. Linear optical perturbation analysis gave insight into the sensitivity of the performance of the interferometers to perturbations of the positions and orientations of the optical elements. Finite element modal analysis yielded structural modes, modeshapes, modal costs, and Hankel singular values. These models were integrated with the disturbance models allowing for generation of frequency response functions. The result of this analysis is end-to-end disturbance characterizations (starlight fringe coherence as a function of reaction wheel speed, for example).

A concept for a space-based interferometer dedicated to the detection of extrasolar earth-like planets is presented. Detection is done in the near infrared (10 micrometers ) where the expected star to planet flux ratio is down to 10⁶ compared to 10¹⁰ in the visible. The longer wavelength also makes is easier to avoid light scatter due to optics micro-roughness. Parent star cancellation is obtained with a rotation shearing interferometer working at its null on axis. The interferometer is of the Fizeau configuration with an aperture composed of twelve 1.2 meter mirrors on a 20 meter ring. This size, which corresponds to a resolving power of 0.05 arcsecond, allows for the investigation of about 30 candidate stars. The interferometer is supported by a chemically rigidized structure deployed by inflation. All optical elements are passively cooled to about 70 degrees Kelvin to reduce the instrumental infrared background. The spacecraft is located at the second Lagrangian point of the earth-sun system in order to minimize attitude control and baffling requirements.

We describe a concept for an orbiting astronomical observatory which will allow high spatial resolution far-UV observations of nearby stars. The scientific goal is to study stellar activity and mass loss using imaging and spectroscopy. Specific areas of study include stellar surfaces, large scale magnetohydroynamic effects, interacting binaries and stellar winds. The instrument is an interferometer with an 8-meter baseline providing 3 milliarcseconds resolution at 1200 Angstrom. The interferometer configuration is of the Fizeau type which affords excellent ultraviolet throughput because of the small number of reflections. The collecting aperture is composed of six 0.6 meter diameter elements distributed on a circle in such a way as lead to near uniform u-v plan coverage when the instrument is rotated around the line of sight. This will lead to excellent imaging capabilities. The interferometer individual channels are kept coaligned and coherent using the light of a nearby guide star. The supporting structure is folded for launch and automatically deployed once on orbit. To minimize disturbance torques and thermal shocks, the spacecraft will be located on a high earth orbit or at the Lagrangian point.

The Orbiting Stellar Interferometer (OSI) is a space-based interferometer concept which will perform high precision wide angle astronomic measurements. The goal of this mission is to achieve an accuracy of 2 uas over a 30 degree field of view. In addition, to astrometry, OSI will also be capable of performing synthesis imaging measurements with a resolution of 14 mas. These capabilities will enable a large number of astrophysics measurements including parallax measurements, dynamics of globular clusters and detection of extra solar planets.

This paper describes the overall design and planned phased delivery of the ground-based Micro-Precision Interferometer (MPI) Testbed. The testbed is a half scale replica of a future space-based interferometer containing all the spacecraft subsystems necessary to perform an astrometric measurement. Appropriate sized reaction wheels will regulate the testbed attitude as well as provide a flight-like disturbance source. The optical system will consist of two complete Michelson interferometers. Successful interferometric measurements require controlling the positional stabilities of these optical elements to the nanometer level. The primary objective of the testbed is to perform a system integration of Control Structure Interaction (CSI) technologies necessary to demonstrate the end-to-end operation of a space- based interferometer, ultimately proving to flight mission planners that the necessary control technology exists to meet the challenging requirements of future space-based interferometry missions. These technologies form a multi-layered vibration attenuation architecture to achieve the necessary quiet environment. This three layered methodology blends disturbance isolation, structural quieting and active optical control techniques. The paper describes all the testbed subsystems in this end-to-end ground-based system as well as the present capabilities of the evolving testbed.

This paper describes the development of the truss structure at the Jet Propulsion Laboratory that forms the backbone of JPL's Micro-Precision Interferometer (MPI) Testbed. The Micro- Precision Interferometer (MPI) Testbed is the third generation of Control Structure Interaction (CSI) Testbeds constructed by JPL aimed at developing and validating control concepts. The MPI testbed is essentially a space-based Michelson interferometer suspended in a ground- based laboratory. This instrument, mounted to the flexible truss, requires nanometer level precision alignment and positioning of its optical elements to achieve science objectives. A layered control architecture, utilizing isolation, structural control, and active optical control technologies, allow the system to meet its vibration attenuation goals. Success of the structural control design, which involves replacement of truss struts with active and/or passive elements, depends heavily on high fidelity models of the structure to evaluate strut placement locations. The first step in obtaining an accurate structure model is to build a structure which is linear.

The Micro-Precision Interferometer Testbed is essentially a space-based Michelson interferometer suspended in a ground-based laboratory. The purpose of the testbed is to serve as a proving ground for technologies needed for future space-based missions requiring low- vibration environments. A layered control architecture, utilizing isolation, structural control, and active optical control technologies, allows the system to achieve its vibration attenuation goals. This paper focuses primarily on the optical design for the testbed and the systems-level tradeoffs between the optics and other systems due to the fact that the interferometer is on a large, lightly damped, flexible structure rather than on the ground. The testbed is designed to be a fully functioning interferometer spacecraft and makes use of flight-like hardware where possible, including an external star simulator, an attitude control system, fringe detection and tracking systems, delay lines, pointing control, laser metrology systems, and computers and electronic subsystems. The engineering decisions that led to the current optical configuration are presented and explained.

This paper describes the validation of an integrated modeling tool using data collected from a laboratory set-up. The modeling package, Integrated Modeling of Optical Systems (IMOS), combines structural modeling, optical modeling and control system simulation into a single environment. The Micro-Precision Interferometer Testbed, a ground-based version of a full- scale spaceborne interferometer, provides the opto-mechanical problem for this investigation. The objective of the effort is twofold: (1) validate the predictive capabilities of IMOS; (2) initiate the controller design for the subsystem under investigation. Ground-based validation of this modeling tool will provide a crucial step towards the ultimate goal of accurately predicting on-orbit behavior of future precision optical instruments.

There seems to be a strong correlation between the number of moving parts on a spacecraft, and the quality and quantity of science that it can be achieved. This is especially true for applications with demanding pointing and alignment requirements like spaceborne interferometry. Unfortunately, moving parts are expensive, and the desire to add moving parts to maximize science conflicts with NASA's current climate of costs constraints. The intent of this paper is to provide the interferometer (or other mission) designer with an overview of the technical issues that confront the cost-effective design and specification of precision spacecraft actuators. First, the paper describes the capabilities and limitations of common actuator components such as bearings, prime movers, and displacement sensors. Next, the paper describes some generic actuator configurations for typical applications. Finally, the paper provides tips on how to write actuator requirements.

The Solar Ultraviolet Network (SUN) is an instrument based on interferometric concepts, and capable of observations with a spatial resolution of 0.013' (10 km) on the Sun, in the UV and visible wavelength ranges. In this paper we present results on fringe pattern acquisition and stabilization as performed on a Mach-Zehnder set up representative of the interferometer cophasing system. The system algorithm is based on 'white light' fringe tracking controlled in a reference interferometer by a synchronous detection. This servo-system drives a two-stages delay line for real-time compensation of the optical path delays. Acquisition capabilities and stability possibilities are investigated as a function of flux and noise levels. Being stabilized, actively cophased, and in a 'compact' configuration, the SUN interferometer possesses remarkable imaging capabilities allowing high resolution diffraction-limited imaging on an extended field of view of 6 X 6 arcsec². The dynamics of reconstructed images is superior to 400 for phase stabilities >= (lambda) /6 and photon flux of approximately 10,000 ph s^-1 pixel^-1 (on average). The SUN instrument is part of the Solar Interferometric Mission for Ultrahigh Resolution Imaging and Spectroscopy (SIMURIS) which was proposed to ESA in the framework of the Next Medium Size Mission (M2) in November 1989, and which completed a First Phase of Study in the context of the Space Station in August 1991.

Heterodyne interferometers have been commercially available for many years. In addition, many versions have been built at JPL for various projects. This activity is aimed at improving the accuracy of such interferometers from the 1 - 30 nanometer level to the picometer level for use in the proposed OSI and SONATA missions as metrology gauges. In the null-gauge configuration, we obtained a precision of 0.6 picometers at time scales of 2,500 seconds. In the relative-gauge configuration, we obtained an accuracy of 3.5 picometers rms in vacuum at time scales of few minutes. As absolute gauge with an accuracy of 10 microns over a distance of 10 meters in under construction.

Space-based astrometric interferometer concepts typically have a requirement for the measurement of the internal dimensions of the instrument to accuracies in the picometer range. While this level of resolution has already been achieved for certain special types of laser gauges, techniques for picometer-level accuracy need to be developed to enable all the various kinds of laser gauges needed for space-based interferometers. Systematic errors due to retroreflector imperfections become important as soon as the retroreflector is allowed to either translate in position or articulate in angle away from its nominal zero-point. Also, when combining several laser interferometers to form a three-dimensional laser gauge (a laser optical truss), systematic errors due to imperfect knowledge of the truss geometry are important as the retroreflector translates away from its nominal zero-point. In order to assess the astrometric performance of a proposed instrument, it is necessary to determine how the effects of an imperfect laser metrology system impact the astrometric accuracy. This paper show the development of an error propagation model from errors in the 1-D metrology measurements through the impact on the overall astrometric accuracy for OSI. Simulations are then presented based on this development which were used to define a multiplier which determines the 1-D metrology accuracy required to produce a given amount of fringe position error.

The Orbiting Stellar Interferometer (OSI) is a space based interferometer capable of making astrometric measurements with 2 - 5 uas accuracy. Typical astrometric observations have one target star within the instrument's instantaneous field of view. The presence of other stars within this field will introduce systematic errors in the measurement. In this paper we present a method of accurately determining the position of multiple stars in the field of view. This method is an extension of the conventional synthesis imaging techniques that will be used for OSI. Astrometric measurements in these crowded field will be extremely useful in studying the dynamics of globular clusters and to make measurements in the direction of the center of the galaxy.

SONATA--Small OSI (Orbiting Stellar Interferometer) for Narrow angle Astrometry with Two Apertures--is a concept for a space based interferometer capable of detecting extra-solar planets. The instrument is an extension of the TOPS-0 interferometer testbed concept which is a ground-based dual feed interferometer, and the space-based OSI concept, which is being studied for the Astrophysics Division of NASA. The SONATA instrument uses a quadruple- feed interferometer which will be capable of measuring fringes on four stars simultaneously within a 10 arcminute field of view. The starlight is collected by two 0.4 meter telescopes separated by 7 meters. The use of common collecting optics results in cancellation of a large number of systematic errors found in multiple baseline designs. The targeted astrometric accuracy for SONATA is 0.5 uas. In the photon-noise limit, this performance can be achieved on 14th magnitude objects by integrating for 4 hours. This level of accuracy will enable detection of nearby Earth type planets. The SONATA design uses a non-deploying structure and will be launched on an Atlas II/Centaur for insertion into a 900 km Sun-synchronous orbit.

Future NASA missions in astrophysics, Earth observation, and solar system exploration that require optical communication, optical and infrared imaging, or high precision astrometric measurements impose very stringent demands for the dimensional stability of precision structures and science instrument components. The objective of this paper is to identify the major mechanisms that influence the dimensional behavior of common optomechanical materials, to identify the mechanisms that are important for the proposed missions with critical dimensional stability requirements, and to compare the mission requirements with state-of-the- art material and measurement technologies. This paper discusses the tradeoffs of passive vs. active means of achieving the dimensional stability requirements. The reduction of power consumption and mass, the reliability improvements as a function of the dimensional stability of the structural materials for a typical interferometer are calculated.

Considerable experience has been gained in the construction of images from ground-based interferometry data. A brief introduction to this work is presented, and some lessons learned from this experience that are relevant to space-based interferometers are discussed.

This paper describes key features of the instrument, spacecraft, and mission design for the Precision Optical INTerferometer in Space (POINTS), which have evolved through studies at the Jet Propulsion Laboratory during the last few years. Design of the flight-system configuration has been driven by several considerations. Since the most ambitious science goals require access to a large portion of the sky most of the time, minimal systematic errors, and a 10-year mission life, a high Earth orbit (higher than 50,000 km) is preferred; the nominal has been taken to be a circular orbit of 100,000-km radius. In order to provide a very uniform thermal environment for the instrument, a solar shield supporting an array of solar cells is mounted on a boom and gimballed along two axes so as to remain pointed at the Sun and to provide constant shade for the entire spacecraft. Silicon cells covering about 85% of the roughly 4.8-m-diameter shield and operating at about 100 deg C could supply sufficient power for a 10-year mission life. A unibody design was selected in which the instrument and spacecraft bus are solidly attached to form a single rigid body. Full pointing freedom for the instrument is provided by articulation of the solar shield about two axes plus roll of the entire spacecraft around the Sun direction. With the high orbit and Sun-facing, geometrically simple spacecraft configuration, the effects of solar radiation pressure--the only significant external disturbance to spacecraft acceleration--can be modeled accurately enough to guarantee no compromise in the accurate velocity determination needed to correct astrometric measurements for stellar aberration.

We present the designs for laser distance gauges to be used in the POINTS instrument, and preliminary performance data. For the target 5 micro-arcsecond astrometric accuracy, we must hold or monitor some critical internal dimensions of the POINTS instrument with 2 picometer (pm, 10^-12 m) accuracy for a few hours. The POINTS architecture makes good use of these gauges, minimizing the number and range of dimensions that must change during operation, and maximizing the similarity of the starlight and metrology measured paths. Gauge designs have been developed for both optical-path-differencing (Michelson) and point-to-point measurements (Fabry-Perot). The Michelson fringes have been measured in a differential (comparison) test; the root-two-point variance (analogous to the Allan variance) in the difference of two measurements over essentially identical 1-meter paths was about 2 pm for averaging times between 40 seconds and 6 hours. A second design for the point-to-point measurements incorporates cornercube retro-reflectors in a resonant cavity. We discuss the new problems anticipated in this design, including the problem of maintaining laser alignment in these point-to-point gauges over the +/- 3 degree range of instrument articulation.

We present the results of two parallel studies of the sensitivity of astrometric measurements to misalignment of each of the major optical assemblies in the POINTS instrument. Tilt and displacement of the optics lead to tilt, displacement, and defocussing of the starlight and metrology beams, giving rise to systematic errors. In one method, we derive analytic expressions for the lowest order dependence of the error on the misalignment, and evaluate them in the present interferometer design. In the second method, we use a commercial numerical ray tracing program to calculate the overall optical path travelled through the misaligned starlight and metrology paths; from those results, our own software determines the dependence of the residual error on the original misalignment. These sensitivities are compared to the analytic results for mutual verification. We also discuss the impact these results have had on the design of the instrument.

Newcomb is a design concept for an astrometric optical interferometer with nominal single- measurement accuracy of 100 microseconds of arc ((mu) as). In a three-year mission life, it will make scientifically interesting measurements of O-star, RR Lyrae, and Cepheid distances, establish a reference grid with internal consistency better than 100 (mu) as, and lay groundwork for the larger optical interferometers that are expected to produce a profusion of scientific results during the next century. With an extended mission life, Newcomb could do a useful search for other planetary systems. The instrument is a highly simplified variant of POINTS. It has three (or four) interferometers stacked one above the other. All three (four) optical axes lie on a great circle, which is also the nominal direction of astrometric sensitivity. The second and third axes are separated from the first by fixed 'observation angles' of 40.91 and 60.51 deg. The fourth axis would be at either 70.77 or 78.60 deg from the first. Each interferometer detects a dispersed fringe (channeled spectrum), which falls on a CCD detector array nominally 8 k elements long and a small number of elements wide. With a nominal baseline length of 30 cm and optical passband from 0.9 to 0.3 microns, the Nyquist limit is reached by a star +/- 21 arcmin from the optical axis. The instrument will be constructed of stable materials such as ULE glass, and have neither internal moving parts nor laser metrology.