Adaptive state estimation for ballistic object tracking with nonlinear model and state dependent process noise

In this paper an adaptive estimation scheme for non-linear system has been characterised where the structure of the process noise of the system depends on the states of the system. A ballistic object tracking problem has been considered here to study the behaviour of such Filters. Adaptive Extended Kalman Filter is used to estimate the position and velocity of the falling object when the process noise covariance of the system is unknown. Several cases are considered for the Monte Carlo simulation to evaluate the performance of such estimators.