A tracking algorithm for both highly maneuvering and nonmaneuvering targets

Presents a modified version of the adaptive interacting multiple model (AIMM) algorithm, to enable tracking of both highly maneuvering and nonmaneuvering targets. The improvements include the addition of an adaptive sampling policy and a different gain calculation for the bias filter. The efficiency of the algorithm has been tested on the benchmark problem of the beam pointing control of a phased array radar. Comparisons with the algorithm utilizing fixed sampling rates are also given. The solutions obtained for the benchmark problem, which has many practical implementation aspects, compare very favourably with previous results.