A Switched IMM Estimator based on the Model Probability Cumulant

Taking into account the disadvantage of the potential loss of accuracy due to the overmodeling at these nonmaneuvering times in the interacting multiple model (IMM) method, an improved IMM algorithm based on the Viterbi technique and the model probability cumulant is proposed. First, select one or two optimal models from the model set to enhance the model utilization rate. Then, judge the variation of the target motion state according to the maneuvering detection mechanism based on the model probability cumulant, and finally switch the different IMM methods. Simulation results show that the proposed method has an enhanced performance in tracking a maneuvering target.