Rebecca Lersch, University of Arizona; Tanner Campbell, University of Arizona; Adam Battle, University of Arizona; Neil Pearson, University of Arizona; Vishnu Reddy, University of Arizona
Keywords: spectral mixing, spectroscopy, spectral modeling
Abstract:
Linear Spectral Mixing for Spacecraft Characterization
Rebecca Lersch1, Tanner Campbell12, Adam Battle1, Neil Pearson4, Roberto Furfaro3, Vishnu Reddy1
1Lunar and Planetary Lab, University of Arizona, AZ
2Department of Aerospace and Mechanical Engineering, University of Arizona, AZ
3Department of Systems and Industrial Engineering, University of Arizona, AZ
4Planetary Science Institute, AZ
Abstract
We present a summary of our spectral mixing pipeline to aid in spacecraft characterization. After analyzing telescopic visible wavelength spectra of several spacecraft, we have been able to demonstrate preliminary capabilities to use spectroscopy and linear spectral mixing in spacecraft identification. To obtain our spacecraft spectrum, we use the Robotic Automated Pointing Telescope for Optical Reflectance Spectroscopy (RAPTORS I) located at the University of Arizona. It was built in 2017 by University of Arizona engineering students. It is a 24-inch Newtonian telescope, equipped with a R~30 spectral resolution transmission grating enabling us to do slitless visible wavelength spectroscopy (0.4-1.0 microns). Phase angle variations affect the telescopic spectrum of spacecraft. Phase angle is the angle created by the sun, the target, and the observer on earth. The angle can be from 0°, which is a fully illuminated object, to 180°, which is not illuminated. These changes in phase angle are observed in the object’s spectrum in the form of variations in spectral slope and absorption band depth. When analyzing the spectrum of the spacecraft, we have the issue of non-resolved object characterization. We do not have the capabilities to determine what aspects of the spacecraft are solar panels, metals, or mylars. The spacecraft appears as one source, and we have no information about the shape of the object, or the size. The size of an object is not correlated to the brightness of the object, as phase angle and material albedo can change the observed brightness. To get information about the spacecraft we can use linear spectral mixing, to identify the endmembers and their abundances. Linear spectral mixing is the process of linearly combining the endmember spectra in different ratios to identify the mixture that best correlates to an observed telescopic spectrum. We do this to determine the abundances of the spectral endmembers within the telescope spectrum. This allows us to identify features of spacecraft to aid us in characterization. The characterization allows us to be able to uniquely identify the spacecraft. We use a Python pipeline that we developed to perform the spectral mixing analysis with our chosen endmembers. This pipeline takes in a list of possible endmembers as well as the observed telescope spectrum, and returns the ratio of possible endmembers, and the Mean Squared Error of that mixture. The goal of this project is to create a pipeline that can match observed visible spectra (0.4-1.0 microns) of spacecraft with their associated materials and percent composition. To do this, we use laboratory measured spectral data of paints, mylars, assorted metals, and solar panel materials. We have done modeling to several spacecraft. One we have run our pipeline on the spectra observed of communication satellite (27400) Astra 3A, we have been able to match the satellite to the best combination of materials. This spin-stabilized satellite is largely composed of solar panels, which acts as a single-material-dominated use case for the spectral mixing process. We have developed the preliminary capabilities to classify spacecraft using spectroscopy and spectral modeling, demonstrated on the Satellite (27400) Astra 3A. We plan to continue developing the spectral mixing pipeline for further analysis of spacecraft, and expanding our material database, as well as investigate phase angle effects on the spectral mixing of high phase angle telescope observations.
Date of Conference: September 19-22, 2023
Track: Satellite Characterization