Abstract: Based on a constant acceleration motion law represented by a polynomial, this article proposes a novel dynamic model of maneuver target—polynomial prediction model (PPM) and its optimal filtering algorithm by means of modeling the target state by a polynomial prediction filter. The analytical results show that any target motion represented by the polynomials can be modeled by this PPM since it does not require any prior knowledge of the target dynamics, and that the optimal filtering algorithm corresponding to it can track any maneuvering motion of a target. The simulation results of the maneuvering target tracking verify the effectiveness of the proposed model and algorithm.

Key words: polynomial prediction, dynamic models, maneuvering target tracking, Kalman filter, interacting multiple model