Interferometric Sensor Measuring
and Modeling Environment


To develop and adapt cutting edge software techniques for the analysis of optical and infrared interferometer data, applicable to the evolution of interferometers from ground-based prototypes and testbeds, to the operational ground-based and space-based science instruments such as the Keck Interferometer, the Space Technology 3 Formation Flying Interferometer (ST3), and the Space Interferometry Mission (SIM).


The existing software base for optical and infrared interferometry is largely targeted to engineering staff for instrument performance analysis and customized modules written by astronomers for specific science goals. Very little has been created to help visualize the data for the more general audience or even for astronomers unfamiliar with this type of data. The current thrust of this task is to adapt and/or develop more generalized, highly visual, tools based on the needs of operations staff and the science community for use at the Palomar Testbed Interferometer (PTI), the Real Time Interferometer Control System Testbed (RICST), the Formation Interferometer Testbed (FIT) and SIM System Testbed 3 (STB-3). These tools ultimately will serve as prototypes for mission instrument analysis tools and will help define the minimal telemetry content needed for evaluating the validity of the science return on missions. With a little forethought the output of these tools could provide visualizations suitable for the public and educational outreach. The content of the tools and displays are a collaboration of both scientists and engineers on PTI, Keck, SIM and ST-3.


Specific accomplishments have been in the areas of creating a web-based interface to existing or specially-adapted software for observation planning, real time monitoring (from a science perspective), post data-collection analyses, baseline estimation, siderostat pointing, and illustration of interferometry operations such as the fringe search animation.

Pticat_sm.jpg1) Observation planning - a planning tool called "getCal" was adapted to web-based input by creating a java applet called "PtiApplet" (click image at right to enlarge). Given an observation period and set of targets the user can select from a list of calibrator stars. The output is an ascii script of the observations showing instrument constraints. It is also a direct input to PTI's sequence computer. For futher details see   From this page select pti_getCal.html to get the following worksheet.

99238WL_sm.jpg2) Real time monitoring tools - The program "rtvis", and an associated java applet "PlotVisApplet", both developed in this task , is a variance of the program "vis" used for calculating visibilities for observations at PTI. It outputs 10 diagnositc aids for determining the validity of the scientific observation. Prior to implementation of rtvis, vis was only run in batch mode after completion of nightly observations, and thus, problems that affected the science results could not be diagnosed until after the fact. This implementation allowed the night observer the ability to change parameters or orders of observations in real time. For further details see PTI Postprocessing Services at The image on the left is from Visibility Plots link (click to enlarge). The rtvis help documentation is also available on this page

3) Post data-collection analyses - A web page was implemented as a front end to "vis" to review the results of the previous nights observations remotely. "PlotVisApplet" serves as the front end. For further details see Select Endnight Report.

4) Baseline estimation tool - The baseline of an interferometer (separation between the 2 telescopes) determines its resolving power. The baseline orientation on the source gives you the position in fourier space (the so-called uv-plane). Each night a program "bFit" is run to check the stability of the baseline. The output is retrievable by a web page.

BL_sid_sm.jpg5) Siderostat modeling tool - The siderostats are flat mirrors which bring the starlight into the interferometer. The siderostats are on an alt-az mounting and their axes of revolution are not exactly at the center or surface of the siderostat. An 8-parameter pointing model is fitted every so often. The output of this is available on a web page. For further details see Select the Endnight Report - put in a day number such as 99238 in the fourth box and hot "List" button. The data is best explained in the image on the right (click to enlarge).

Fringetrack_sm.jpg6) Fringe Search animation - One of the more challanging goals of operating an optical or infrared interferometer in space is locating the white light fringe. The white light fringe cannot be seen all at one time due to the way the instrument is designed and constantly jitters about over 10s of nanometers due to moving parts and in ground-based interferometers, atmospheric distortions. The width of the fringe is measured in 10s of nanometers somewhere along the delay line path of the instrument that varies in length from 10s to 100s of meters for the various missions. There are various ways of computing estimators of the fringe location and various search algorithms employed to find it. This animation has been submitted to


This Task...

    This task's goal is to develop operations and science analysis tools for use on the interferometer testbeds as they evolve to the more flightlike or operational versions. PTI is of particular interest since it is a completely functional infrared interferometer that can provide science data that can be exploited to provide provide visualizations suitable for the public and educational outreach as the missions prepare for operational status.

The Instrument...

    An optical or infrared interferometer is a complex state machine that runs in a number of operational modes. It is composed of a number of subsystems which interact in real time feedback control loops. State flags are cleared and set based upon tolerances on each performance loop. Nominally you operate it much in the same way as a standard telescope, you provide a direction to point to, an integration time and a time window for the observation to be taken. However, operationally there the similarity ends. You don't get a picture - you basically get one fourier pixel of the object with an amplitude and phase measurement. (It is possible to get enough such pixels to reconstruct an image but it takes a long time). Interferometers are not scheduled so much as they are sequenced. Scientifically valid data taking only occurs when the state flags of each subsystem are set properly for the operating mode. This means that a 2 minute integration could take 5 or more minutes in elapsed time since the total integration is built up of many small snippets (10s of milliseconds) of valid integrations.

The Data...

    To the scientist astronomical interferometers provide a phase (delay line position) and amplitude (visibility) measurements at a given baseline separation and orientation on stars. In the ideal case, the visibility of a star is related to its structure. A point source will give high visibility while an extended source such as a giant star, a dust enshrowded star, or a close binary star will give low visibility. The phase provides a star's position with respect to nearby reference stars. The phase information between two or more separate point sources in the sky for a mission like SIM will allow for accurate narrow angle astrometry, up to 1 microarcsecond relative postioning, to a set of stable reference stars. This is 1000 times better positional accuracy than the European Space Agency's Hipparcos satellite's determinations. Successive remeasurements over time can yield an "astrometric jitter" which is the signature of low-mass companions such as planets. This simple interpretation of interferometery data is complicated by instrument performance imperfections, and for ground-based interferometers, atmospheric effects. Data base managment aids in keeping track of a stupendous variety data samplings but the science analysis tools for this data tend to be developed by a particular astronomer for specific scientific or engineering goals and for a particular interferometer.

Status/Future Plans

  • The above 6 tasks are complete.
  • Current: (1) A software module called "rtdualstarvis" and its associated java applet will provide a real time display for the dual star astrometry mode. It is due to be completed by December which is the end of the current observing season for PTI.
  • Current: (2) A software module called "ptiObs" which will allow the remote scientist to replay observations taken previously by other observers.
  • Future: New animations to illustrate the principles of operation of interferometers and to visualize interferometry data in ways for the non-interferometrist to understand what the data means.

Points of Contac

Raymond J. Bambery

Eric M. De Jong

Jet Propulsion Laboratory

Jet Propulsion Laboratory

(818) 354-2887

(818) 354-0302



1. The Palomar Testbed Interferometer.
2. The Space Interferometry Mission.
3. Space Technology 3 Mission.
4. Keck Interferometer.
5. Optical Long Baseline Interferometry News .