An attitude determination algorithm for fast maneuver satellites

The nonlinearity of the dynamic and kinetic model becomes so high that ordinary attitude determination algorithms are not accurate enough when the satellite rotates fast, and sometimes the system state does not converge. In this paper, a new attitude determination algorithm is presented to deal with this situation. Based on minimal model error (MME) principle, the model error caused by the nonlinearity is considered and estimated in the algorithm. The approximate model derived from the Taylor formula is modified using the estimation of model error, thus the accuracy of attitude estimation is improved. Simulation results indicate that the algorithm presented in this paper performs better than ordinary EKF algorithm when the satellite is rotating fast.