A Lightweight, Inexpensive, Low-Power Star Tracker for Small Satellites

"Space vehicles designed to resupply systems such as SBIRS Low and Space-Based Laser, as well as maneuverable targets and interceptors, require an affordable, low system-impact sensor for attitude determination, especially when a high degree of navigationis required for high delta-v maneuvers. Today's star trackers are too massive, expensive, and power hungry to appropriately serve small maneuverable vehicles. However, lower-impact technologies, such as sun sensors, do not offer the flexibility andperformance required for advanced mission profiles. AeroAstro and MIT propose an innovative approach to a low-impact star tracker serving these small maneuverable vehicles, balancing accuracy with power consumption, mass, and cost. No solution currentlyexists in this area of the trade space. The design uses a pinhole in place of traditional optics, takes advantage of the processing capabilities embedded in newer CMOS imagers to reduce power consumption and limit the amount of glue logic required, andexploits highly compact pattern recognition algorithms to find star pairs using a minimal star catalog. The solution offers accuracy better than 100 arc-seconds, meeting the unique requirements of small maneuverable space vehicles at a fraction of thecost, mass, and power consumption of larger, higher impact star trackers. AeroAstro and MIT see a wide variety of commercial and scientific applications enabled by the development of this star tracker technology. The system occupies a unique niche abovesun sensors in capability but below CCD-based star trackers in cost and system impacts. The small satellite and maneuverable space vehicle communities have been seeking a component such as this for years, and commercialization of this product is expectedto be received very positively. Ancillary benefits of this development work include an increased understanding of the use of a pinhole to replace bulky traditional optics, advances in developing fast and compact algorithms for pattern recognition, and animproved ability to take advantage of the imb