Improved data association for space-object identification using gating methods and planar triangles
MetadataShow full item record
This thesis develops and establishes a validation gate-based approach that enables one to improve the accuracy of a planar triangle method (PTM) for identification of resident space object (RSOs). It has been shown that RSOs can be identified successfully using data association methods such as the planar triangular method. However, the PTM's performance degrades as the measurement error in the RSOs' position increases. The PTM was developed for spacecraft attitude determination by comparing properties of planar triangles with stars in the field of view (FOV) to that of stars in a catalog. However, it has been shown that the PTM also provides a consistent success rate in RSO identification. The work presented herein proposes the use of gating as a data association method for RSO identification. This thesis shows that accuracy of existing data association methods such as the PTM can be significantly improved by combining gating with the PTM. The gating approach is used to associate the measured observation to its truth. The use of elliptical gates as a precursory step to the PTM shows that the accuracy of the algorithm is improved significantly. The performance of the gating approach as a standalone data association method for RSO identification is also studied. The work done here shows the change in performance trend when the gating-based approach is subjected to various measurement errors. Also, the performance of the gating method in a cluttered environment is studied and compared to the performance of the PTM in a cluttered environment.