Star Identification Algorithm Based on Multi-dimensional Features and Multi-layered Joint Screening for Star Sensors

The algorithm for star identification is a crucial technology determining the orientation of spacecraft using sensors. Traditional algorithms achieve matching by seeking unique or few optimal solutions. However, in high-noise environments, some solutions may be lost, which could result failure. A new lost-in-space architecture aimed at rapid under high position noise conditions directly positions final proposed this paper. main idea to construct sufficiently redundant navigation triangles, fully utilizing physical relationships features and forming screening method from low dimensions loose strict. During identification, multi-layer joint adopted screen triangles as whole, narrowing range matches quickly while retaining error tolerance. In series simulation experiments, achieved rates 99.51%, 99.06%, 98.42% 2.0 pixel noise, 1.0 Mv magnitude 5 false stars, respectively. terms practical application, all 1000 images taken sensor orbit have been successfully identified, it only takes 28ms identify each image. addition, consumer-grade cameras ground also show that has strong robustness interference more severe environments. This provides partial algorithmic reference non-specialized design sensors low-cost, large-scale satellites future.