Modeling and Simulation of a LongWave Infrared Polarimetric

Modeling and Simulation of a Long-Wave Infrared Polarimetric Sensor for Space Object Detection and Characterization Kevin Pohl

Introduction SSA Today • Ground and space sensors, both private and government owned/operated • Radars • Telescopes • Voluntary sharing of telemetry data, two-line elements, etc. , between organizations that maintain space object catalogs • Coverage and responsiveness are limited by geographical location of sensors • Most optical sensors are visible spectrum and are further constrained by local time of day • The concept modeled in this project uses LWIR polarimetry • Short exposure time • Day/night capable Image courtesy of Air University

Polarimetry • Polarization measurement is described by the Stokes vector • Based on six flux measurements using ideal polarizers in front of a radiometer: horizontal (PH), vertical (PV), diagonal (45 and 135 degrees; P 45 and P 135), and left (PL) and right circular (PR) (1) The circular polarization component (s 3) is very small compared to the linear components and can be ignored • The Stokes components are then paired into three unique color spaces • S 1/S 0 • S 2/S 0 • S 2/S 1

Polarimetry • (2)

Model Setup • Two versions of the primary vehicle were created • Vehicle 1 a with a bus covered in Kapton • Vehicle 1 b with a bare aluminum bus • Both versions are simple cubes with a small dish antenna and a single large solar panel Primary vehicles, aluminum bus (left) and Kapton bus (right) • The secondary vehicle has the shape of a 3 U cubesat, but is scaled up to roughly half the size of the primary Secondary vehicle

Results • Comparing Vehicle 1 a (“Baseline”) by itself to Vehicle 1 a with the secondary object nearby results in a similar magnitude of difference as comparing Vehicles 1 a and 1 b (different bus material) to each other

Results • With the secondary object partially obscured the statistical difference is reduced

Results • The aluminum bus emits more strongly polarized light than the Kapton covered bus, resulting in a much smaller difference between the baseline data set and the data with the secondary object in the scene

Conclusion & Future Work • This method allows the discrimination between a known object and that same object with something else nearby • A sufficient amount of the secondary object must be visible in order to detect a difference • This method is better suited to LEO applications than GEO • In GEO the scene diversity is severely limited by the nature of the orbits • In LEO the secondary object will be visible from various angles over the course of multiple passes over a ground site • Future work: • Further refine the simulation to incorporate more materials and shapes • Determine the optimum number/distribution of ground sensors to provide good coverage of LEO

Aerospace and Ocean Systems Laboratory Representative Programs Autonomy and Resilience Squad. X Core Technologies & Experimentation Autonomous control of EM battlespace to combat adversary UAS platforms Hallmark TA 1: Tools/Technology SOSI, proximity operations, and tracking Cognitive orchestration, robust/secure control Maritime Remote Sensing SOSI situational awareness in degraded or denied sensing environments NAVAIR Airworthiness Center Cybersecurity in airworthiness, failure modes, criticality, and common criteria Code 31 Low-SWAP EW Payloads Physics-based models, computational fluid dynamics November 2018 Satellite ground station and cubesat launches Introduction to the Hume Center Low-power GPU platforms for cognitive EW missions, leveraging SDR and deep learning 10