A Novel Improved Cubature Kalman Filter with Adaptive Generation of Cubature Points and Weights for Target Tracking
Hongpo Fu,Yongmei Cheng,Cheng
DOI: https://doi.org/10.1088/1361-6501/ac3785
IF: 2.398
2021-01-01
Measurement Science and Technology
Abstract:In nonlinear state estimation, the generation method of cubature points and weights of the classical cubature Kalman filter (CKF) limits its estimation accuracy. Inspired by this, in this paper, a novel improved CKF with adaptive generation of the cubature points and weights is proposed. Firstly, to improve the accuracy of the classical CKF while considering the calculation efficiency, we introduce a new high-degree cubature rule combining the third-order spherical rule and the sixth-degree radial rule. Next, in the new cubature rule, a novel method that can adaptively generate cubature points and weights based on the distance between the points and center point in the sense of the inner product is designed. We use the cosine similarity to quantify the distance. Then, based on that, a novel high-degree CKF (HCKF) is derived that uses much fewer points than other HCKFs. In the proposed filter, based on the actual dynamic filtering process, the simultaneous adaptive generation of cubature points and weight can make the filter reasonably distribute the cubature points and allocate the corresponding weights, which can obviously improve the approximate accuracy of the one-step state and measurement prediction. Finally, the superior performance of the proposed filter is demonstrated in a benchmark target-tracking model.