Single observer passive locating is an important task for air reconnaissance. A robust algorithm for air-borne single observer passive locating based on phase difference is proposed in this paper. The target location can be worked out by using the Kalman filter model based on the azimuth angle of target signal and the phase changing difference at the same time. The Kalman filter model for calculating the phase difference rate of change is presented. And the Interaction Multiple Model(IMM) filter model is also given for raising the precision of target locating algorithm. In the experiment, the STK software is utilized for generating the platform position data, the signal azimuth data and the phase changing difference. In addition, the target locating result and the performance are also analyzed. Finally, several constructive suggestions are made for practical engineering applications.