篇名 | Variational Bayesian-based Multiple-model Algorithm Combined with Auxiliary Knowledge |
---|---|
卷期 | 31:2 |
作者 | Jing Liu 、 Gao-De Qin 、 Quan-Hui Wang 、 Xue-zhu Na 、 En Fan |
頁次 | 250-263 |
關鍵字 | ADS-B 、 multiple-model 、 target tracking 、 variational Bayesian 、 EI 、 MEDLINE 、 Scopus |
出刊日期 | 202004 |
DOI | 10.3966/199115992020043102021 |
The theoretical framework of the multiple-model algorithm combined with auxiliary knowledge is first proposed in this paper. In order to improve the tracking accuracy of tracking algorithms that is in situation with unknown measurement noise, where Automatic Dependent Surveillance-Broadcast (ADS-B) equipment is employed to keep track of aircraft, a variational Bayesian-based multiple-model algorithm combined auxiliary knowledge (KAVBMM) is proposed. To solve the state and noise distribution, the variational Bayesian approximation is adopted for performing multiple known distribution approximation and estimate the measurement noise variance. Meanwhile, the KAVBMM algorithm utilizes a multiple-model method to adapt the maneuvering change of the target and adjusts the measurement noise value according to Navigational Accuracy Category (NAC) for position information. The results of simulation experiment and read-data experiment shows that the proposed KAVBMM algorithm is can improve the tracking performance.