T.J. Rodigas, L3Harris Applied Defense Solutions; Kevin Melloy, L3Harris Applied Defense Solutions; Mark Bolden, L3Harris Applied Defense Solutions; Michael Butterfield, L3Harris Applied Defense Solutions; Ryan Shier, L3Harris Applied Defense Solutions
Keywords: Space Situational Awareness, Sensors, Imaging, Telescopes, Photometry
Abstract:
There are many approaches to providing data that can be used for Space Situational Awareness (SSA). A common method is to deploy a world-wide network of task-track telescopes that observe individual objects or object clusters. Such networks are usually limited by two main problems: task-track telescopes typically have a small field of view (FOV), and they usually capture images in only a single bandpass. This means that additional telescopes are necessary to persistently observe and characterize satellites in multiple filters. An alternative, and potentially more cost-effective approach, is to employ an optical network of wide FOV Bayer arrays that simultaneously generate images in three visible spectrum filters. In 2016 L3Harris Applied Defense Solution (ADS) deployed such a network to address this problem for SSA, and we have been maintaining and upgrading the network since then.
The vast majority of digital cameras employ Bayer arrays in order to acquire RGB color images, which are generally consistent with what our eyes see. Bayer array technology is employed across multiple markets, including consumer digital cameras spanning from smartphones to digital single-lens-reflex (SLR) cameras. The industrial market leverages Bayer array technology to permit acquisition of color images for machine vision, security, inspection, and a plethora of applications. Innovation in CMOS sensor technology using Bayer arrays has decreased the noise, increased the sensitivity, and decreased the cost of these sensors; they are rapidly becoming the go-to option for small aperture astrophotography and stellar science. Multiple astronomical research groups have adopted Bayer arrays for cost-effective solutions, and there are published methods for mapping Bayer RGB photometry to astronomical bands such as the Johnson system. Applying these techniques to photometric extraction for satellites presents additional challenges not addressed in the scientific literature.
Since Bayer arrays do not filter light in traditional photometric bands, instrumental photometric magnitudes must be calibrated using standard astronomical star catalogs. There are multiple approaches to this technical challenge, each with their own benefits and drawbacks. A common method is to use published transformation equations, but these rely on calibrated observations of stars whose spectral energy distributions (SEDs) are quite different from satellites. The calibrations are also usually produced from the ground using different telescopes on different nights, which introduces additional uncertainties into the transformations.
Another option is to simply pretend that the Bayer RGB filters are the same as the Johnson BVR filters, since the respective filter profiles are similar. However, this method inevitably suffers from the seemingly minute filter differences, which can contribute to major discrepancies in resident space object (RSO) photometry. Obtaining an accurate estimate of a satellites photometry in a standard photometric system is thus a significant challenge.
A potential solution to this problem is to stop trying to convert Bayer RSO photometry into standard catalog systems and instead convert catalog star photometry from standard systems into the Bayer system. This makes computing RSO photometry a straightforward process once the relevant reference stars photometry have been converted to the Bayer system.
Here we describe our method for producing a complete Bayer RGB star catalog. We outline how we computed Bayer RGB photometry of ~ 2 million stars in our production star catalog. To test the validity of our approach, we computed the aggregate photometric accuracy (on thousands of detected stars) from real images acquired from one of our global network sites. We then compared these results to the other methods that convert Bayer photometry to the Johnson system. Our results are promising, with our Bayer->Bayer approach being up to 2x more accurate.
We will also present initial Bayer RGB results on real satellites observed at one of our sensor sites. We will present a selection of the thousands of multicolor lightcurves we have acquired so far. Finally, based on our more accurate Bayer lightcurves, we will discuss and provide an initial assessment on how Bayer array multicolor photometry might possibly be leveraged to gain insight into a satellites structure, stability, and orbit.
Date of Conference: September 17-20, 2019
Track: Non-Resolved Object Characterization