An adaptive filter for the track prediction of highly maneuvering anti-ship cruise missiles

Abstract

An adaptive filter is developed for the track prediction of anti-ship cruise missiles (ASCM). The scenario is an anti-ship missile defense problem of a ship defending against an incoming ASCM. The filter processes the fire-control radar returns of range, elevation and azimuth information, and it is part of the ship's gunfire control system. A generic ASCM is characterized by a parametric model of propulsion, aerodynamics and guidance. The parameters cover drag, lift, thrust and four guidance parameters. An adaptive extended Kalman-Bucy filter is derived to estimate the parameters and the states of the model. Three of the parameters are actually step functions. Process noise is adaptively added as a function of measurement residuals to allow the filter to track the time-varying parameters. The adaptive filter has been tested on a class of six ASCM types. They include subsonic seaskimmers with end game pop-up maneuvers and supersonic high altitude divers. Good estimates of the model parameters were obtained roughly 4-6 seconds after the start of a maneuver. The adaptiveness of the newly developed filter follows rapid ASCM maneuvers after settling into steady-state. The probability of kill criterion is used to compare the performance of the new tracking filter with two filters that are currently in the U.S. Fleet.