Abstract: The abrupt faults of electromechanical actuator (EMA) can cause the safety problem of plane flight control system, so as to the plane lose control. A diagnostic method for abrupt faults of electromechanical actuator (EMA) was proposed based on interactive multiple model (IMM) and unscented Kalman filter (UKF). The UKF was used to make the state equation be close to a nonlinear feature and make the filter represent a better stability and an acceptable computational complexity. And the IMM was used to solve the problem of the diagnosed model with a few measurable parameters, but also to improve the speediness and accuracy of the fault diagnosis in the case of large difference between the occurrence of the fault and the presupposed fault. The effectiveness of proposed IMM-UKF fault diagnosis method of EMA was verified by simulation. The results show that the proposed IMM-UKF method can obtain real-time and accurate diagnostic values of the abrupt faults of components and sensor in EMA.