Adaptive filtering and imaging algorithms for tracking maneuvering targets

摘要

Evasive target maneuvers create problems for traditional tracking algorithms because maneuver accelerations are frequently abrupt and not adequately modeled by wideband noise processes. Earlier work has suggested the use of an imaging sensor to determine target orientation, from which the likely direction of a maneuver acceleration can be inferred. The authors compare this strategy with other, more traditional approaches in the context of tracking a maneuvering target in the plane. The augmented measurement filter, EKFp (extended Kalman filter-p), achieved excellent performance results in both nominal operation and in operation during the maneuver, indicating the superiority of the image augmented filtering system. This superiority in performance comes at the cost of additional hardware and a more complicated filter, but these indicate that this augmented measurement filter is a fruitful area of inquiry.