Interacting multiple model algorithm based on S-amended UKF

Devoted to the problem of maneuvering target tracking under nonlinear observation, an S-amended unscented Kalman filtering (SUKF) is developed in this paper, which uses the idea of S-amended anti-divergent method for Kalman filter in linear filtering and improves the performance of unscented Kalman filtering algorithm. Then based on the coordinated turn (CT) models, adopting the method of SUKF, an interacting multiple model (IMM) algorithm named interacting multiple model algorithm based on S-amended unscented Kalman filtering (IMM-SUKF) is researched. Simulation results show that this algorithm can effectively improve the tracking precision of the multiple model algorithm, especially when the model mismatching and the target´s sudden maneuvering occurs, and that it is suitable for engineering applications, especially for tracking highly maneuvering aerial targets.