Combined with the existence of transiently lost reference signals and outside disturbance， this paper targets the problem on trajectory tracking control of unmanned aerial helicopter （UAH）. Firstly， the lost reference signals in short time interval are divided into two parts including stationary reference signals and non-stationary ones. Secondly， as for the lost signals， a prediction model based on differential auto-regression integration moving average （ARIMA）model is utilized to predict the stationary signals and the Markov analytic method is used to predict the non-stationary one. Thirdly， by combining the predicted data and tracking error， two slide mode controllers are proposed for the transverse-longitudinal subsystem and the vertical subsystem in the UAH system， which can greatly improve the tracking performance and eliminate the negative effects of the outside disturbance. Moreover， the proposed control algorithm is further improved to decrease the chattering effect induced by the sliding mode. Finally， simulations are presented to illustrate the effectiveness and advantages of the derived control schemes.