Research Of Marginal Distribution Bayesian Filter In Turn Maneuvering Targets Tracking

Target tracking is to estimate the states of different targets by using the measurements obtained by a sensor at different times.It has wide applications in military and civil fields,therefore is of important research significance.The multi-target Bayesian filter based on the finite set statistics theory(FSST)has become a main research topics in the multi-target tracking field because it avoids the data association which is required in traditional multi-target tracking method.It has been shown that the multiple model method is efficient for tracking the maneuvering target with unknown maneuvering mode.However,the existing multiple model method may result in the loss of the target track and inaccurate target state estimation because the difference between the actual target motion mode and the models used in a filter.Aiming at this problem,the research on the method for tracking the turn maneuvering target with unknown and variable turn rate is presented in this thesis by applying the estimation of turn rate and turn direction angle to the marginal distribution multi-target Bayesian(MDMB)filter.The main content of this thesis is summarized as follows:1)We introduce the multi-target Bayesian filter based on the FSST,probability hypothesis density(PHD)filter,MDMB filter and Gaussian mixture PHD filter for jump Markov system models(JMS-GMPHD filter).2)We propose the estimate method of turn rate for the turn maneuvering target in the two-dimensional space.Applying the turn rate estimation into the MDMB filter,we present the MDMB filter with turn rate estimation for tracking the maneuvering target moving in the two-dimensional space.Based on the simulation experimental data,we analyze the tracking performance of the proposed filter by comparing it with the JMS-GMPHD filter.3)We propose the estimate method of turn rate and turn direction angle for the turn maneuvering target in the three-dimensional space.Applying the turn rate estimation and turn direction angle estimation into the MDMB filter,we present the MDMB filter with turn rate estimation and turn direction angle estimation for tracking the maneuvering target moving in the three-dimensional space.Based on the simulation experimental data,we also analyze the tracking performance of the proposed tracking filter by comparing it with the JMS-GMPHD filter.