The interacting multiple model algorithm based on adaptive Markov transition probability

In view of problems of the current statistical (CS) model for weak maneuvering targets tracking, this paper combines it with constant velocity (CV) model, using the interacting multiple model (IMM) algorithm to estimate target states. The traditional algorithm with fixed transition probability matrix is improved by using an adaptive method, and it can adjust transition probabilities according to the measured data of each moment automatically. The simulation results show that, whether strong or weak maneuvering targets, the new algorithm has better tracking performance than the traditional IMM algorithm.