KEYWORDS: Signal to noise ratio, Telescopes, Device simulation, Convolutional neural networks, Education and training, Diffraction, Formation flying, Stars, Monte Carlo methods, Error analysis
Starshades are a leading technology to enable the direct detection and spectroscopic characterization of Earth-like exoplanets. To maintain high contrast during observations, the starshade and telescope must keep within 1 m of relative alignment over large separations (>20,000 km). This formation flying is made possible with precise spacecraft position information obtained through accurate sensing of the occulted star’s diffraction peak (referred to as the spot of Arago) incident on the telescope aperture. We present a lightweight image processing method based on a convolutional neural network paired with a simulation-based inference technique to estimate the position of the spot of Arago and its uncertainty. On simulated images, the method achieves an accuracy of a few centimeters across the entire telescope aperture. By deploying our method at the Princeton Starshade Testbed, we demonstrate that the neural network can be trained on simulated images and used on real images and that it can successfully be integrated in the control system for closed-loop formation flying.
Princeton's starshade testbed has been utilized extensively over the past 5 years towards satisfying key milestones of NASA's "Starshade to TRL5 (S5) Technology Development Plan.”1,2,3,4 The initial optomechanical design & build of the long-travel (76m) laser testbed was presented at SPIE AS16 (2016).5 Since then, several key optomechanical upgrades have proven crucial to final milestone completion, including various light-tighting measures, thermal insulation & stabilization, cleanliness measures, a motorized X-Y camera stage, and—especially—a novel, light-tight, low-profile, multi-mask changer (for remote-operable toggling between various starshade masks, without disturbing the stability of the testbed's interior thermal and cleanliness environments). These final upgrades are summarized herein.
Following the results of Starshade Milestone 1 in which we demonstrated broadband contrast better than 10−10, we have performed model validation experiments to show that diffraction models accurately predict the contrast due to perturbations representing the building blocks of the instrument error budget. The perturbations include the displacement of petal edge segments, sinusoidal petal edge shape deformations, global petal position errors, and random petal radial placement errors. We also show that the model accurately predicts the combined effects of two errors. The experiments result in a measured Model Uncertainty Factor (MUF) that is then applied to the starshade instrument contrast error budget.
The Astro2020 report recommended that NASA’s next flagship have direct imaging of rocky, Earthlike exoplanets as its core science mission. At present, the starshade is the only high contrast imaging technique that has demonstrated broadband contrast at levels suitable for imaging exo-Earths in the laboratory. A starshade is an occulter positioned to cast a shadow of an exoplanet’s host star onto the telescope aperture, and narrow enough that the nearby exoplanet remains visible to the telescope. The starshade has a precise shape tailored to suppress diffraction of the starlight into the shadow. Starshade-based observations also have other advantages compared to coronagraphic observations. These include: no effective limit to the outer working angle, higher throughput for the exoplanet light, dramatically simpler requirements on the telescope optics, and the ability to provide high contrast at ultraviolet wavelengths. These advantages come at the price of needing a separate spacecraft to fly the starshade in formation with the telescope, and the consequent costs in fuel and time required for stationkeeping and retargeting. We describe work being done to mature starshade technology to technology readiness level 5 (TRL 5) in NASA’s S5 activity. This work includes optical measurements of a starshade’s ability to suppress light at levels required for a flagship mission, laboratory demonstrations of position sensing and control methods for starshade formation flying, and manufacture and test of flight-like starshade mechanical assemblies that can deploy accurately and stably to the precise shape required for starlight suppression in space.
Starshades are a leading technology to enable the direct detection and spectroscopic characterization of Earth-like exoplanets. Critical starshade technologies are currently being advanced through the S5 Project and at the Princeton starshade testbed. We report on the conclusion of Milestone 2 of the S5 Project, optical model validation. We present results from optical experiments of starshades with intentional perturbations built into their design. These perturbations are representative of the type of perturbations possible in a flight design and serve as points of validation for diffraction models and error budgets. We show agreement between experiment and diffraction model that meets the Milestone 2 criteria of 25% agreement. We then place these results into the larger context of the design and error budget of a full scale starshade mission. We also present the latest updates to the development of non-scalar diffraction models relevant to the testing of sub-scale starshades. This work completes the optics-focused S5 technology milestones that put starshade technology at TRL 5.
This special section is dedicated to starshades: science, engineering, technology, and programmatics. Our reasons for organizing this special section are several fold. First as a new technology and with research accomplished in many institutions, recent results are widely scattered in the literature. As such, we see great value in colocating many of the most recent results. This guest editorial summarizes the 19 contributed papers as the result of a special call for papers. Since this is a rapidly maturing technology, we wanted to colocate a primer with the most current work in the field. It is hoped that this primer will provide a tutorial to the starshade concept and pathway to the literature not in this section. In doing so, we hope to widen the starshade community in terms of engineering and scientific engagements. This tutorial takes the form of a dialogue, where frequently asked questions are answered.
KEYWORDS: Signal detection, Signal to noise ratio, Photon counting, Venus, Planets, Image processing, Sensors, Point spread functions, Electron multiplying charge coupled devices, Exoplanets
Because exoplanets are extremely dim, an electron multiplying charge-coupled device operating in photon counting (PC) mode is necessary to reduce the detector noise level and enable their detection. Typically, PC images are added together as a co-added image before processing. We present a signal detection and estimation technique that works directly with individual PC images. The method is based on the generalized likelihood ratio test (GLRT) and uses a Bernoulli distribution between PC images. The Bernoulli distribution is derived from a stochastic model for the detector, which accurately represents its noise characteristics. We show that our technique outperforms a previously used GLRT method that relies on co-added images under a Gaussian noise assumption and two detection algorithms based on signal-to-noise ratio. Furthermore, our method provides the maximum likelihood estimate of exoplanet intensity and background intensity while doing detection. It can be applied online, so it is possible to stop observations once a specified threshold is reached, providing confidence for the existence (or absence) of planets. As a result, the observation time is efficiently used. In addition to the observation time, the analysis of detection performance introduced in the paper also gives quantitative guidance on the choice of imaging parameters, such as the threshold. Lastly, though our work focuses on the example of detecting point source, the framework is widely applicable.
KEYWORDS: Planets, Signal detection, Venus, Exoplanetary science, Signal to noise ratio, Point spread functions, Photon counting, Telescopes, Coronagraphy, Image processing
A starshade suppresses starlight by a factor of 1011 in the image plane of a telescope, which is crucial for directly imaging Earth-like exoplanets. The state-of-the-art in high-contrast post-processing and signal detection methods was developed specifically for images taken with an internal coronagraph system and focused on the removal of quasi-static speckles. These methods are less useful for starshade images where such speckles are not present. We are dedicated to investigating signal processing methods tailored to work efficiently on starshade images. We describe a signal detection method, the generalized likelihood ratio test (GLRT), for starshade missions and look into three important problems. First, even with the light suppression provided by the starshade, rocky exoplanets are still difficult to detect in reflected light due to their absolute faintness. GLRT can successfully flag these dim planets. Moreover, GLRT provides estimates of the planets’ positions and intensities and the theoretical false alarm rate of the detection. Second, small starshade shape errors, such as a truncated petal tip, can cause artifacts that are hard to distinguish from real planet signals; the detection method can help distinguish planet signals from such artifacts. The third direct imaging problem is that exozodiacal dust degrades detection performance. We develop an iterative GLRT to mitigate the effect of dust on the image. In addition, we provide guidance on how to choose the number of photon counting images to combine into one co-added image before doing detection, which will help utilize the observation time efficiently. All the methods are demonstrated on realistic simulated images.
Starshades are a leading technology to enable the detection and spectroscopic characterization of Earth-like exoplanets. We report on optical experiments of sub-scale starshades that advance critical starlight suppression technologies in preparation for the next generation of space telescopes. These experiments were conducted at the Princeton starshade testbed, an 80-m long enclosure testing 1/1000’th scale starshades at a flight-like Fresnel number. We demonstrate 10 − 10 contrast at the starshade’s geometric inner working angle (IWA) across 10% of the visible spectrum, with an average contrast at the IWA of 2 × 10 − 10 and contrast floor of 2 × 10 − 11. In addition to these high-contrast demonstrations, we validate diffraction models to better than 35% accuracy through tests of intentionally flawed starshades. Overall, this suite of experiments reveals a deviation from scalar diffraction theory due to light propagating through narrow gaps between the starshade petals. We provide a model that accurately captures this effect at contrast levels below 10 − 10. The results of these experiments demonstrate that there are no optical impediments to building a starshade that provides sufficient contrast to detect Earth-like exoplanets. This work also sets an upper limit on the effect of unknowns in the diffraction model used to predict starshade performance and set tolerances on the starshade manufacture.
Starshades are a leading technology to detect and characterize Earth-like exoplanets. In this paper we report on optical experiments of sub-scale starshades that advance critical starlight suppression technologies in preparation for the next generation of space telescopes. These experiments were conducted at the Princeton starshade testbed, an 80 m long enclosure testing 1/1000th scale starshades at a flight-like Fresnel number. In this paper we summarize recent updates made to the starshade testbed and optical model. We present results from recent experiments testing two starshade masks with intentional perturbations built into their shape. One of the perturbed masks has three petals that are shifted radially outward by 7-11 microns and the other mask has two petals shifted radially outward plus two petal edge segments displaced from their nominal position. We show the model agrees with experiment to better than 25% accuracy. These results are placed into context with previous experiments on perturbed shapes and progress made towards satisfying a critical milestone in advancing starshade technology to TRL 5.
One of the most promising mission concepts for imaging and characterizing Earth-like planets about other stars is a telescope/starshade system. But, in order to maximize the number of targets accessible for a given mission formation-keeping strategies are required. This paper details the design and implementation of a formation ying hardware-in-the-loop simulation in the existing Princeton Starshade Testbed to validate formation sensing and control algorithms while maintaining high contrast with a ight-like starshade. We present initial experimental results to help advance the starshade technology gap of formation sensing and control.
We present an analysis of the Rayleigh scattering in the Princeton starshade testbed and show that it explains several notable features in the contrast images. The scattering is consistent with that expected due to air molecules and does not require airborne dust to explain. Rayleigh scattering limits the observable contrast at the ~ 1 × 10-11 level at the inner working angle in the contrast images, but it limits the observable suppression at ~ 10-9 level. We present a crude estimate of the level of scattering of starlight to be expected in a flight starshade due to zodiacal dust in the solar system and conclude that it is unlikely to be observable. We comment on whether Rayleigh scattering drives longer starshade testbeds to operate in vacuum.
Starshades are a leading technology to enable the direct detection and spectroscopic characterization of Earth-like exoplanets. Starshade starlight suppression technology is being advanced through sub-scale starshade demonstrations at the Princeton Starshade Testbed and we present here the successful completion of a technology milestone focused on the demonstration of high contrast at flight-required levels. We demonstrate 10-10 contrast at the inner working angle of a starshade with a flight-like Fresnel number at multiple wavelengths spanning a 10% bandpass. We show that while contrast at the inner working angle is limited by the presence of non-scalar diffraction as light propagates through narrow slits between the starshade petals, high contrast is still achieved over most of the image. Successful completion of this milestone verifies we can design a starshade capable of producing scientifically useful contrast levels.
Starshades provide a leading technology to enable the direct detection and spectroscopic characterization of Earth-like exoplanets. Two key aspects to advancing starshade technology are the demonstration of starlight suppression at science-enabling levels and validation of optical models at this high level of suppression. These technologies are addressed in current efforts underway at the Princeton Starshade Testbed. Recent experimental data suggest we are observing the effects of vector (non-scalar) diffraction, which are limiting the starshade's performance and preventing the scalar optical models from agreeing with experimental results at the deepest levels of suppression. This report outlines a model developed to simulate vector diffraction in the testbed using a full solution to Maxwell's equations propagating through narrow features of the starshade. We find that experimental results can be explained by vector diffraction as light traverses the thickness of the starshade mask and that our model is in rough agreement with observations. We provide simulation results of a number of starshade geometries as a first attempt to understand the relation of these effects to properties of the starshade masks. Finally, we outline a number of possible solutions aimed to minimize vector effects and to allow us to reach our milestone of 10-9 suppression.
A starshade is a promising instrument for the direct imaging and characterization of exoplanets. However, even with a starshade, exoplanets are difficult to detect because detector noise, starshade defects, and misalignment (dynamics of the starshade system) degrade the signal to noise ratio (SNR) and contrast. No image processing methods have been specialized for images produced by a starshade system (simply referred as starshade images later). In this paper, we present a method, based on the generalized likelihood ratio test (GLRT), to detect and characterize planets from a single starshade image or multiple starshade images. This paper describes the GLRT model and its preliminary results for simulated images with starshade shape error, dynamics, detector noise and starshade rotation considered. The planets are detected with low false alarm rate, and planet positions are accurately estimated, and planet intensities are reasonably estimated. Thus, it demonstrates great potential as an acute and robust detection method for starshade images
Direct imaging using a starshade is a powerful technique for exoplanet detection and characterization. No current post-processing methods are specialized for starshade images and the ones for coronagraph images have not been applied to images produced by a starshade system ( starshade system means the light sources, starshade and telescope). Here, we report on the first step towards adapting these methods for starshade systems. We have built a starshade imaging model. We generate the image based on a simulation of the real astronomical scene and consider the effects of various starshade defects, misalignment, wavefront error, and detector noise. Future work will add the system dynamics of formation flying between the starshade and the telescope. The ultimate goal is to adapt coronagraphic image processing methods for starshade imaging.
Starshades are a leading technology to enable the direct detection and spectroscopic characterization of Earth-like exoplanets. Two key aspects to advancing starshade technology are the demonstration of starlight suppression to the level required for flight and validation of optical models at this high level of suppression. These technologies are addressed in current efforts underway at the Princeton Starshade Testbed. We report on results from modeling the performance of the Princeton Starshade Testbed to help achieve the milestone 10−9 suppression. We use our optical model to examine the effects that errors in the occulting mask shape and external environmental factors have on the limiting suppression. We look at deviations from the ideal occulter shape such as over-etching during the lithography process, edge roughness of the mask, and random defects introduced during manufacturing. We also look at the effects of dust and wavefront errors in the open-to-atmosphere testbed. These results are used to set fabrication requirements on the starshade and constraints on the testbed environment. We use detailed measurements of the manufactured occulting mask to converge towards agreement between our modeled performance predictions and the suppression measured in the testbed, thereby building confidence in the validity of the optical models. We conclude with a discussion of the advantages and practicalities of scaling to a larger testbed to further advance the optical aspect of starshade technology.
A starshade is a specially designed opaque screen to suppress starlight and remove the effects of diffraction at the edge. The intensity at the pupil plane in the shadow is dark enough to detect Earth-like exoplanets by using direct imaging. At Princeton, we have designed and built a testbed that allows verification of scaled starshade designs whose suppressed shadow is mathematically identical to that of space starshade. The starshade testbed uses a 77.2 m optical propagation distance to realize the flight Fresnel number of 14.5. Here, we present lab result of a revised sample design operating at a flight Fresnel number. We compare the experimental results with simulations that predict the ultimate contrast performance.
The external starshade is a method for the direct detection and spectral characterization of terrestrial planets around other stars, a key goal identified in ASTRO2010. Tests of starshades have been and continue to be conducted in the lab and in the field using non-collimated light sources. We extend the current approach to performing night-time observations of astronomical objects using small-scale (10-30cm) starshades and the McMath-Pierce Solar Telescope at Kitt Peak National Observatory. This configuration allows us to make measurements of stars with a Fresnel number close to those expected in proposed full-scale space configurations. We present the results of our engineering runs conducted in 2015.
The external starshade is a method for the direct detection and spectral characterization of terrestrial planets
around other stars, a key goal identified in ASTRO2010. In an effort to validate the starlight-suppression performance of
the starshade, we have measured contrast better than 1×10-9 using 60 cm starshades at points just beyond the starshade
tips. These measurements were made over a 50% spectral bandpass, using an incoherent light source (a white LED), and
in challenging outdoor test environments. Our experimental setup is designed to provide starshade to telescope
separation and telescope aperture size that are scaled as closely as possible to the flight system. The measurements
confirm not only the overall starlight-suppression capability of the starshade concept but also the robustness of the setup
to optical disturbances such as atmospheric effects at the test site. The spectral coverage is limited only by the optics and
detectors in our test setup, not by the starshade itself. Here we describe our latest results as well as detailed comparisons
of the measured results to model predictions. Plans and status of the next phase of ground testing are also discussed.
KEYWORDS: Astronomy, Telescopes, Solar telescopes, Observatories, Electron multiplying charge coupled devices, Stars, Space telescopes, Mirrors, Control systems, Calibration
The direct detection and characterization of an Earth-like exoplanet is of the highest scientific priority and a leading technology that will enable such discovery is the starshade external occulter. We report on the latest results in ground-based efforts for demonstrating and advancing the technology of starshades. Using the McMath- Pierce Solar Telescope at the Kitt Peak National Observatory, we are able to track stars as they move across the night sky and stabilize a beam of starlight behind a starshade. This has allowed us to conduct the first astronomical observations achieving high-contrast with starshades. In our latest efforts, we have extended the separation between the starshade and telescope to reach an inner working angle of 10 arcseconds at a flight-like Fresnel number and resolution. In this report, we detail the development of a closed-loop feedback system to further stabilize the beam at the extended baseline and provide results on the contrast achieved. We conclude by laying out future work to design a dedicated siderostat-starshade facility for future testing of and observations with starshades. Our main result: we achieved a broadband contrast ratio of 3:2 x 10-5 at 15 arcseconds IWA, while at a flight-like Fresnel number and resolution.
We report on laboratory demonstrations of a vision-based sensor to aid in the formation flying of suborbital vehicles. Precision formation flying of such vehicles will allow us to hold a starshade external occulter in the line of sight between a telescope and star at large separations. This will enable us to perform the first astronomical demonstrations of starshades as we attempt high-contrast imaging of the outer planetary systems of nearby stars. In this report, we identify two sensor architectures and detail the equations for a closed loop visual feedback system to be used for precision formation flying. We investigate the sensor's expected performance through a suite of Monte Carlo simulations and system-level demonstrations in the lab. We also report on the development and demonstration of a means for visual attitude and position determination.
We review the progress on the New Worlds Airship project, which has the eventual goal of suborbitally mapping the Alpha Centauri planetary system into the Habitable Zone. This project consists of a telescope viewing a star that is occulted by a starshade suspended from an airship. The starshade suppresses the starlight such that fainter planetary objects near the star are revealed. A visual sensor is used to determine the position of the starshade and keep the telescope within the starshade’s shadow. In the first attempt to demonstrate starshades through astronomical observations, we have built a precision line of sight position indicator and flew it on a Zeppelin in October (2012). Since the airship provider went out of business we have been redesigning the project to use Vertical Takeoff Vertical Landing rockets instead. These Suborbital Reusable Launch Vehicles will serve as a starshade platform and test bed for further development of the visual sensor. We have completed ground tests of starshades on dry lakebeds and have shown excellent contrast. We are now attempting to use starshades on hilltops to occult stars and perform high contrast imaging of outer planetary systems such as the debris disk around Fomalhaut.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.