A tracking method for GEO space targets based on adaptive memory attenuation UKF

Abstract

Aiming at the problem that the nonlinearity of the tracking state equation and the problem of strong noise in the measurement process in the process of tracking the target, a non-cooperative target tracking method in geostationary earth orbit based on adaptive memory attenuation UKF is proposed. Firstly, the relative motion model of the non-cooperative target is constructed, and the state equation and measurement equation are established. Use the memory attenuation adjustment factor to enhance the correction effect of the new measurement on the state estimation, improve the adaptability of the filtering method to nonlinearity and noise, and finally the relative motion trajectory of the space non-cooperative target is obtained. Finally, the standard UKF algorithm and improved UKF algorithm are applied to target tracking and compared by simulation. The deviation of the three coordinate axis components of the actual motion trajectory and the predicted trajectory of the space target is used as the evaluation index. The simulation results show that the filtering accuracy of the improved UKF algorithm is significantly improved, while taking into account the real-time performance of on-orbit tracking.