Improving bearings-only target state estimation tracking problem by using adaptive and nonlinear kalman algorithms

Abstract

Finding the best estimate of the process state from noisy data is the main problem in tracking systems, many efforts and researches have been done to remove this noise. More useful information about the target’s state can be extracted from observations by using a more appropriate model for the target’s motion or using additional sensors. In this paper, we will introduce two methods to improve the estimation of bearing-only target tracking problem in two dimensions (2D). The first method is by adding a third sensor and making a good alignment of those sensors, and at the same time an extended Kalman filter (EKF), unscented Kalman filter (UKF) and cubature Kalman filter (CKF) are implemented. The second method is by applying an adaptive nonlinear Kalman filter (ANKF) for two sensors to solve the problem of measurement variance uncertainty.