An Angle-Parameterized Hierarchical-Splitting Gaussian Sum Based Tracking Method Using One-Dimensional Cosine Angle and Doppler Rate-of-Changing by a Single Satellite

摘要

In order to achieve high precision tracking of a moving airborne emitter by a single satellite platform, a novel passive tracking method, combined one-dimensional cosine angle of arrival (CAOA) and Doppler rate-of-changing, is presented. To solve the high nonlinearity in the measurements, an efficient tracking algorithm called the angle-parameterized hierarchical-splitting Gaussian sum cubature Kalman filter (APHSGSCKF) is also proposed. The APHSGSCKF algorithm consists of multiple parallel CKFs. They are initialized by a proposed method called orthogonal cosine angle parameterized hypotheses (OCAPH), which divides the area target possible lying using a parameterized cosine angle. And the area is determined by a noised one-dimensional CAOA at the beginning of the measurement. In the splitting and merging conduct, to reduce the amount of computation while improving the accuracy, a new metrics of the degree of nonlinearity is introduced. The value range of it is segmented hierarchically, which determines how many filters should be split into. All the filter outputs are used to estimate the motion state. Simulation results show that the proposed algorithm can reach better performance compared with the general single CKF algorithm and the Gaussian sum cubature Kalman filter (GSCKF) initialized by the proposed OCAPH method.