Interacting multiple model (IMM) algorithm is one tracking filtering algorithm for the target that has a variety of motion state models. Motion state changes and motion model switches can be represented by homogeneous Markov Chain, Target filtering results is weighted by multiple filtering model results, in order to achieve an effective target trajectory tracking. Currently, theoretics research on target trajectory tracking is focused on tracking model selecting and tracking algorithm. This paper designs one combination model of constant and turning motion to track filtering by interacting multiple model algorithm, sequentially it achieve the tracking and estimation of target trajectory. Simulation results demonstrate the effectiveness of the proposed method.