Target tracking technology has been widely applied in both military and civil areas, thus many countries and researchers have paid much more attention. The core part of the target tracking technology is filtering algorithm. It is a hotspot and nodus in target tracking technology research to propose more reliable filtering algorithms to cope with the non-linear and non-Gaussian problem, and applied in practical tracking system efficiently.Firstly, some basic principles of target tracking and mathematical models of moving targets are introduced in this paper, and analyzed these models' application field, advantages and disadvantages.Secondly, based on the Gaussian Filter (GF), we focus on some nonlinear filtering methods including Extend Kalman Filter (EKF), Unscented Kalman Filter (UKF) and Cubature Kalman Filter (CKF). Simulations using target tracking model designed in the paper prove the validity of each nonlinear filtering algorithm. The advantages and disadvantages of the algorithms are analyzed and compared.In non-linear and non-Gaussian target tracking systems, the GF which tracking precision decreases or diverges, cannot satisfy application request. The Particle Filter (PF) that provides a solution of approximate bayes for arbitrary non-linear and non-Gaussian systems, has important theoretical and practical value. In this paper, the basic principle and implementation steps of PF are described, and the improved Particle filter algorithm based on importance density function is studied.In existing Gussian Particle Filter (GPF), sample importance density function is constructed through combining the latest measurement based on GF. However, the true posterior probability density could be approximated badly by the GF under the condition of high accuracy, strong nonlinearity measurements. In order to solve this problem, a Truncated Adaptive Cubature Kalman Filter (TACKF) is proposed, based on which a new sample importance density function is constructed, so that a Truncated Adaptive Cubature Particle Filtering (TACPF) method can be derived. Simulation results show that the proposed filtering algorithm has higher estimation accuracy than existing GPF for addressing the nonlinear state estimation with high accuracy and strong nonlinearity measurements. |