In view of the accurate tracking for the airport surface targets, a multi-sensor trajectories fusion method based on geography guidance is proposed to process the data from Automatic Dependent Surveillance – Broadcast (ADS-B) and Multilateration (MLAT). This method deals with the relations among the different sensors using distribute filters to realize the trajectories fusion. The process of filtering is under the control of the airport surface geography information and improves the precision through a variable structural multi-model mechanism. Experimental results show that the proposed method has the advantages of both high detection precision and project application value.