The improved algorithm of UPF in spacecraft autonomous optical navigation

Abstract

An autonomous optical navigation scheme for cruise phase is presented using the star light data and body edge data which is measured by star sensor and optical navigation camera. And the relative position for probes is acquired from the star light data and body edge data. The observation information is pretreated by wavelet analysis and filter the high frequency noise, then uses the wavelet to recompose the signal to gain smooth parameters value and makes a process of UPF (unscented particle filter) based on these work, so as to reduce the variance of importance weights. The real time orbit for probes is determined by using these. The arithmetic integrates wavelet analysis and UPF to improve the precise and reliability of the autonomous optical navigation system. Simulation result shows that the improved UPF algorithm is not only more accurate but also has higher rate of convergence compared with the UPF.