Research On Maneuvering Target Tracking Algorithm Based On Variable Structure Multiple-Model

Maneuvering target tracking is a process of modeling, estimation and tracking for maneuvering targets by using a variety of known information and measurement data.Maneuvering target tracking technology is widely used in military field and civil field.With the development of science and technology, the maneuverability of targets keeps enhancing, which causes the much more difficulties in tracking maneuvering target.Therefore, it is of great significance to research the technology of maneuvering target tracking.This thesis mainly studies the variable structure multiple model estimation method in the applications of single maneuvering target tracking and multiple maneuvering target tracking. Firstly, the basic theory of multi-model estimation method and the basic principle of the interactive multiple-model estimation algorithm are introduced, particularly. Through a simulation experiment, the performance of several commonly used multiple model tracking algorithm is analyzed and compared.Secondly, in the light of the problem that homogeneous Markov transition probability matrix is commonly used in variable structure multiple model estimation, the posteriori estimation of the Markov transition probability matrix is adopted and strong tracking interacting multiple model algorithm is used to improve the adaptive ability of multiple model estimation in this thesis. Thirdly, a variable structure multiple model estimation algorithm which the best model is augmented based on the minimum model group is proposed. The algorithm effectively reduces the computational complexity and improves the tracking accuracy compared with the traditional optimal model augmented estimation algorithm. Finally, aiming at the problem of multiple target tracking, the interactive multiple model joint probabilisticdata association(IMMJPDA) filter algorithm is studied, and on this basis, a kind of variable structure multiple model joint probabilistic data association(VSMMJPDA)filter algorithm is proposed.Variable structure multiple model estimation method is another milepost of multiple model estimation method. It overcomes many shortcomings of fixed structure multiple model estimation method and meets the modern requirements of multiple model estimation; so it has broad prospects for development. At the end of this thesis, four outlooks are given aiming at the shortcomings of the existing and development directions of variable structure multiple model estimation and pointed out the direction for further research in the future.