A single Kalman filtering algorithm for maneuvering target tracking

Abstract

Maneuvering tracking involves tracking a target that is suddenly accelerated by the pilot. Two approaches have been proposed for maneuvering target tracking: the residual-based approach and the multiple-model approach. The latter approach provides better tracking performance but has higher computational complexity compared with the former one. The authors present a new maneuvering tracking algorithm which combines the benefits of both approaches so that the tracking accuracy is comparable with that of the multiple-model approach and the complexity with that of the residual-based approach.< >