Attitude Model Identification of On‐orbit Satellites

Lihui Geng,Tao Zhang,Deyun Xiao,Jingyan Song
DOI: https://doi.org/10.1108/00022660910941848
2009-01-01
Abstract:PurposeThe purpose of this paper is to propose an identification algorithm to obtain generalized attitude model (GAM) of satellites in on‐orbit environment, which includes missing attitude data and multi‐noise. The identified GAM and noise model are the basis of attitude control and state estimation on‐orbit.Design/methodology/approachTo cope with noises contaminating both input and output of attitude model, the errors‐in‐variables model is transformed into a traditional Box‐Jenkins model according to the attitude control loop. The wavelet denoising (WD) technique is helpful to predict the missing output data using the identified GAM.FindingsBy the numerical simulation, it is verified that the proposal accompanied with WD has a faster prediction capability than that of the algorithm without WD. As a result, the proposed approach is suitable to attitude model identification of on‐orbit satellites.Originality/valueThis identification algorithm can deal with two kinds of on‐orbit conditions and has a fast parameter convergent rate. Therefore, it has a practical application value in on‐orbit environment.
What problem does this paper attempt to address?