Application of a two-stage adaptive extended Kalman filter for satellite attitude control systems delay and faults estimation

Abstract

This paper presents applications of a two-stage adaptive extended Kalman filter (TSAEKF) strategy to satellite attitude control systems (ACSs) input delay, actuator and sensor faults estimation simultaneously. A general nonlinear system model with input delay is established based on the separate-bias principle. Two ACSs system models with input delay and unknown actuator/sensor faults are established. Then, based on two-stage extended Kalman filter (TSEKF) and adaptive fading Kalman filter (AFKF), a TSAEKF strategy for states/delay/faults estimation of nonlinear system is presented. Simulation results of a satellite ACSs with input delay and actuator/sensor faults indicate that the TSAEKF strategy can estimate the system states and the amplitude of input delay and faults directly and rapidly.