Robust Adaptive Target State Estimation for Missile Guidance using Interactive Multiple Model Kalman Filter
Small Business Information
500 West Cummings Park, Suite, 3000, Woburn, MA, 01801
Raman K. Mehra / Constant
AbstractThe accuracy and robustness of target state estimators to target maneuvers and varying target characteristics have been one of the limiting factors to improving the interceptor's terminal performance and direct hit capability. This effort will examine techniques to improve the accuracy of the onboard target state estimators, and thereby improve the miss distance performance of homing interceptors. The solution proposed here to improve tracking performance is based on the Interacting Multiple Model (IMM) Kalman filter estimator which has been used successfully by Scientific Systems for ship-board radar target tracking of highly maneuvering targets under ECM and clutter conditions. The IMM Kalman filter can handle target maneuvers with a significantly smaller estimation error and track loss rate compared to a single Kalman filter. The IMM-KF has proven to be robust and extremely fast in maneuver detection. We will consider both active and passive tracking of targets by the interceptor. The active tracking problem is similar to the one considered earlier, but the passive or semi-active LOS tracking problem is non-linear and poorly observable. We will use an approach by Jauffret and Pillon (1996) to linearize the measurement equation, and investigate the effects of using an accurate initial position estimate obtained by a radar. We will also investigate the use of sparse active measurements and their effect on the tracking performance. The combined performance of an Adavance Guidance Law (AGL) and IMM-KF algorithm will be evaluated using the simulation of a navy interceptor missile. Prof. Y. Bar-Shalom, a leading authority in the field of estimation and target tracking and one of the developers of the IMM-KF algorithm will provide support to the project as a consultant.
* information listed above is at the time of submission.