Pose angle aided 3D target tracking models are derived based on the analysis of target velocity variation under the yaw and pitch angle, with the relationship between the pose angle and the motion. Aiming to the non-Gaussian characteristic of pose angle measurements, the Gaussian sum square-root cubature Kalman filter(GSSCKF) is proposed based on the analysis of square-root cubature Kalman filter(SCKF), and the filtering ability of non-Gaussian nonlinear is improved. According to the pose angle variation law during target motion, tracking models with different pose angle components are established, and the maneuvering pose angle is estimated by model switch. The simulation results show the effectiveness and correctness of the tracking model and the filtering algorithm.