Volume Kalman Filter Algorithm For Target Tracking Performance Analysis And Research

Nonlinear filtering is a very important aspect in target tracking.In this paper,the volume Kalman filter algorithm and its improved algorithm in target tracking are studied deeply.Firstly,the background and significance of the research are given,and the research status of the target tracking and filtering methods is briefly introduced.Then,the principle and process of the classical nonlinear filtering algorithm are summarized based on the Bayesian filtering framework.On this basis,the main research results of this paper are given.First,an adaptive volume Kalman filter(ADCKF)algorithm with recursive estimation of noise characteristics was proposed to solve the problem of target tracking and noise covariance estimation of nonlinear systems when the noise characteristics are unknown.This algorithm is based on the adaptive filtering of constant noise estimation statistics,combining the covariance matching principle with the noise recursive estimation method,and then substituting it into the volume Kalman filter(CKF)algorithm.The improved algorithm does not depend on the prior noise distribution.It can estimate and modify the statistical characteristics of noise on-line according to the observation information,and verify the effectiveness of the algorithm through the simulation experiment of the conventional point target tracking.Secondly,an improved iterative adaptive volume Kalman filter was proposed to solve the navigation problem of collision avoidance.Through the simulation experiment of ship target tracking,it is proved that the improved algorithm has strong robustness and improves the filtering accuracy and stability of the traditional CKF algorithm.In the process of target estimation,the new algorithm can reduce the error of observation data and track and estimate more accurate target data.The relationship between the range of parameters and the performance of filter estimation in the improved algorithm is also studied.Thirdly,in view of the influence of different parameter values on the performance of CKF algorithm,this paper studies the influence of CKF algorithm on the performance of filtering estimation when the five main parameters of initial value of state variable,initial covariance matrix,state transition matrix,process noise covariance and measurement noise covariance are different,so as to get the best selection range of these parameters.