An improved fuzzy alpha-beta filter for tracking a highly maneuvering target

In this paper, a new fuzzy alpha-beta filter for tracking a highly maneuvering target is presented. The target may maneuver at uniform angle rates by executing sharp turns within a short time. A set of fuzzy if-then rules is used to estimate the index of maneuverability (tracking index λ). This index λ is used to estimate the optimal αβ filter coefficients. The proposed filter is referred to as the Improved Fuzzy alpha-beta filter (IFαβ filter). It is shown through numerical simulations that the proposed filter does not only achieve a good tracking accuracy, but can also be used to detect the start and the end of target maneuvers. Compared to the standard fuzzy alpha-beta filter, the proposed fuzzy algorithm is better, in terms of both accuracy and computational complexity.