IMM-CKF for a Highly Maneuvering Target Using Converted Measurements

We consider the state estimation of a highly maneuvering aircraft using an air moving target indicator (AMTI) radar. The AMTI radar measures the range, azimuth, and radial velocity of the target. In our previous work, we used the interacting multiple model (IMM) filter with the extended Kalman filter (EKF), unscented Kalman filter (UKF), and cubature Kalman filter (CKF) using the AMTI measurements. In this paper, we use the converted measurements in an IMM-CKF and analyze the performances of these two types of algorithms. The converted measurements include the unbiased converted measurement (UCM) and modified unbiased converted measurement (MUCM) or conditional mean. The performances of the filters are analyzed using the root mean square position and velocity errors, root time-averaged mean square (RTAMS) error, average normalized estimation error squared (ANEES), mode probabilities, and computational cost. We also compute the posterior Cramér-Rao lower bound to evaluate these two types of algorithms.