Neuromorphic and Optical Constellation Tracker for Unstable spacecraft Attitude correction (NOCTUA)

Star trackers are an important sensor on a satellite used to accurately determine the attitude of the spacecraft. Attitude determination require position, velocity, and time (PVT) information. A satellite can be unaware of its attitude due to any number of factors including deployment issues or hardware/software errors, this loss of positional awareness is known as lost-in-space. To acquire attitude information and solve the lost-in-space problem, star trackers are used to measure the positions of stars relative to the spacecraft and compare the pattern of the observed stars to their known absolute position from a star catalog. Existing star tracking attitude sensors are limited in their ability to operate above an angular rate of ~3-5 degrees/second, additionally as CubeSats and nanosats become more popular, there is a growing need for a low power, non-resource intensive attitude determination method that is capable of operating at higher angular rates than existing star trackers. Neuromorphic or event-based cameras appear to be ideal for solving these needs. Prime Photonics proposes to develop the Neuromorphic and Optical Constellation Tracker for Unstable spacecraft Attitude correction (NOCTUA) event-based star tracker. Event cameras are shutterless, and only detect changes in pixel intensity from frame-to-frame, not complete images. As a result, they offer much greater dynamic range at much lower data rates, processing power, and cost, than their conventional camera counterparts. The lower data rates and processing power make them an ideal sensor for resource limited spacecraft. Prime has experience with event cameras in prior R&D efforts including an event camera-base system to monitor debris ingestion at the inlet of a military aircraft engine. In this prior work we have developed code for multi object detection, velocity vector determination, and centroid calculation of event objects. For the solicitation application, Prime has already completed desktop proof-of-concept testing using an event camera rotated by hand up to 150 degrees per second and was able to detect star patterns printed on a test sheet, where a traditional frame-based camera’s output image was too blurry to resolve the stars. In the Phase I, we will demonstrate the ability of the NOCTUA system to track stars and determine its attitude while under high rates of angular spin and compare its performance to a traditional star tracker. We will modify an existing star tracking algorithm to work with events, as well as determine the event camera’s sensitivity limit (i.e., how faint can a star be to be useful in attitude determination). In Phase II, more advanced features of the system will be demonstrated, and the prototype device will be matured for use in military and commercial applications.