Orbit determination via adaptive Gaussian swarm optimization

Accurate orbit determination (OD) is vital for every space mission. This paper proposes a novel heuristic filter based on adaptive sample-size Gaussian swarm optimization (AGSF). The proposed estimator considers the OD as a stochastic dynamic optimization problem that utilizes a swarm of particles in order to find the best estimation at every time step. One of the key contributions of this paper is the adaptation of the swarm size using a weighted variance approach. The proposed strategy is simulated for a low Earth orbit (LEO) OD problem utilizing geomagnetic field measurements at 700 km altitude. The performance of the proposed AGSF is verified using Monte Carlo simulation whose results are compared with other advanced sample based nonlinear filters. It is demonstrated that the adopted filter achieves about 2.5 km accuracy in position estimation that fulfills the essential requirements of accuracy and convergence time for OD problem.