Star spot location estimation using Kalman filter for star tracker

Abstract

Star pattern recognition and attitude determination accuracy is highly dependent on star spot location accuracy for the star tracker. A star spot location estimation approach with the Kalman filter for a star tracker has been proposed, which consists of three steps. In the proposed approach, the approximate locations of the star spots in successive frames are predicted first; then the measurement star spot locations are achieved by defining a series of small windows around each predictive star spot location. Finally, the star spot locations are updated by the designed Kalman filter. To confirm the proposed star spot location estimation approach, the simulations based on the orbit data of the CHAMP satellite and the real guide star catalog are performed. The simulation results indicate that the proposed approach can filter out noises from the measurements remarkably if the sampling frequency is sufficient.

Angular velocity estimation from measurement vectors of star tracker

“…7, there are large variation and uncertainty for the least-squares estimation given by Eq. (12), especially in the roll axis. Obviously, the designed Kalman filter increases the accuracy remarkably, and convergence is achieved after about 90 s (about 60 images).…”

“…Acquire the stars appearing in both the previous and current frame by the recursive star centroiding and identification algorithms under the tracking mode (see [12] for details), and then mark them as k, k 1; 2; …; n.…”

“…In this paper, we develop a method to estimate the angular velocity by measurement vectors of the star tracker in real time, and an adaptive Kalman filter is designed on different guide stars in a same-star image frame without information of the dynamic model and inertia dyadic. Actually, this paper integrates the work presented in [12], where a Kalman filter approach has been implemented to improve the star acquisition procedure.…”

The Motion Planets Detection and Tracking Algorithm Based on Gestalt Principle

“…But there is a large amount of calculation data and poor real-time performance. Liu et al [5] use the likelihood function and target motion model to design Kalman filter tracking algorithm, but it is unable to track multiple targets simultaneously. Barniv et al [6] build the dynamic programming model to avoid speed mis-match problem by optimization principle, but this algorithm fails when the optimal principle does not meet.…”

Real-time estimation of ship's horizontal attitude based on horizon tracking

“…To the best of our knowledge, we are the first to estimate the ship's horizontal attitude based on the horizon information. In the horizon detection approach, which is motivated by [16], particular attention is paid to the use of a predicted ROI with the designed Kalman filter. As the horizon detection is only performed in small windows, the proposed approach takes a short time and performs robust to the false alarms.…”