Robust adaptive target state estimation for missile guidance using the interacting multiple model Kalman filter

In this paper we present an application of the IMM algorithm to maneuvering target state estimation from line of sight (LOS) and range/range-rate measurements (semiactive case). These target state estimates are used by the homing loop (guidance law) to guide the interceptor towards the target. Standard terminal homing guidance laws, such as proportional navigation, have proven to be very effective against non-maneuvering targets, but their performance degrades drastically in the presence of maneuvering targets. Modern guidance laws have been developed to try to improve terminal homing (miss distance) performance in the presence of maneuvering targets. However, these new guidance laws requires precise target state estimation, and the accuracy and robustness of the standard target state estimators to target manoeuvres have proven to be one of the limiting factors on improving the interceptor's terminal performance. The IMM Kalman filter in this case was designed with a set of three target models: one for the non-maneuvering situations, one for mild maneuvering situations, and one for high maneuvering situations. Two different measurement rate updates were used in this problem: one for LOS measurements and one for range/range-rate measurements. This approach was tested in seven different scenarios (provided by NSWC) using a seeker simulator developed by APL/JHU. In all cases, the estimator was able to get good estimates of target's velocity and acceleration, improving homing performance (for 2 of the targets we compared the IMM versus a single Kalman filter, and in both cases the IMM performed better).