Parameter Adaptive Tracking Algorithm Based on Maneuver Detection

In maneuvering target tracking， the traditional current statistical model Kalman filter algorithm has low tracking accuracy for weak / inorganic maneuvering target， significantly reduces the tracking accuracy for sudden maneuvering， and the tracking performance is limited by prior parameters. To solve the above problems， this paper proposes a parameter adaptive maneuvering target tracking algorithm based on maneuvering detection. The algorithm uses the probability distribution characteristics of residual to construct double threshold detection threshold， and adjusts the parameters adaptively according to the detection results. Firstly， by using the variance information of acceleration prediction error， the maneuvering frequency and acceleration variance are adaptively adjusted to overcome the problem of prior setting of model parameters and improve the tracking accuracy of weak maneuvering target； Secondly， the fading factor is introduced after the detection of maneuver， which makes the introduction time of fading factor more reasonable and enhances the response ability of the algorithm to maneuver. The simulation results in two typical maneuvering scenarios show that the proposed method can better adapt to acceleration step maneuver and turning maneuver than the Kalman filter algorithm based on the current statistical model with fixed parameters.