Comparison Between Nonlinear Filtering Techniques for Spiraling Ballistic Missile State Estimation

Abstract

During the reentry to the atmosphere, certain ballistic missiles are known to undergo violent spiraling motions induced by aerodynamic resonance between roll and yaw/pitch modes. Successful interception of such spiraling targets is critically dependent on the performance of the target state estimator. Strong nonlinearities involved in the system dynamics and measurement equations together with sensor noise make this a challenging estimation task. The performance of an extended Kalman filter (EKF), an unscented Kalman filter (UKF), and a particle filter (PF) designed for this estimation problem is compared in this paper. Additionally, a hybrid Rao-Blackwellized PF (RBPF) approach combining the EKF and the PF is also considered. Simulation results are provided to support the conclusions from the present study.