The standard Kalman filter algorithm cannot accurately preprocess the measured data of aeroengine with exceptional errors. The principle of standard Kalman filter and the impact of test errors to the filter estimate results were analysed, and the method of dynamically adjusting the weight of observation information in the filter estimate result was introduced. Then, based on M-estimation theory, the Robust Kalman filter principle and the recursion formula were presented. The state-space equations and observation equations of the measured parameters were established in terms of CA(Constant Acceleration)model. In order to decrease the calculation consumption, the sequence filter was applied separately to process the different sensed data. Furthermore, the preprocessing to the simulation sensed data of a given turbofan engine's steady operation was carried out as an example, using the given Robust Kalman filter. The calculation results, compared with standard Kalman filter, show that the designed Robust Kalman filter has better estimate precision with a given model error.