Target Tracking Simulation Based on Maximum Likelihood Kalman Filtering

Abstract

The Maximum Likelihood Kalman Filtering (MLKF) has some problems in the target tracking prediction stage, which is mainly shown as the calculation amount is large and the calculation accuracy is not high in the prediction stage. In order to solve the above problems, an improved MLKF algorithm is proposed by combining an Iterative Approximation algorithm (IA) with MLKF. In each iteration, the iterative approximation algorithm can solve an approximate quadratic programming problem with linear constraints to obtain an analytical solution to the approximate optimization problem. By using this iterative approximation algorithm to solve the Maximum Likelihood (ML) estimation, more accurate results can be obtained. Monte Carlo simulation results show that using iterative approximation method to solve ML estimation in the prediction stage can make MLKF have higher tracking accuracy.