Two-stage identification method for attitude-control models of on-orbit satellites

A two-stage identification method was developed for attitude-control models of on-orbit satellites for accurate attitude control. The attitude-control models are based on a generalized attitude model (GAM) and its corresponding noise model (NM). These are both low order models which are able to characterize the main satellite dynamics and the corresponding noise. These low-order attitude-control models facilitate improved controller design and state estimation. This identification approach uses two open-loop identification processes, with higher order auxiliary models used in the estimation for filtering and with the filtered signals used to identify the low order GAM and NM. An identification experiment with a micro-satellite simulator was used to verify the effectiveness of the two-stage identification method.