华中科技大学学报（自然科学版）2017，Vol.42Issue(2)：89-94,6.DOI:10.13245/j.hust.170217
基于多模型GGIW-CPHD滤波的群目标跟踪算法
Group targets tracking algorithm using a multiple models Gaussian inverse wishart CPHD filter Group targets tracking algorithm using a multiple models Gaussian inverse wishart CPHD filter
摘要
Abstract
Gamma Gaussian inverse wishart probability hypothesis density (GGIW-CPHD)filter algo-rithm was always used to track an unknown number of group targets and variable measurement rates in the presence of clutter measurements and missed detections.Aiming at the defect that the tracking error of GGIW-CPHD algorithm has greatly increased in the group maneuvering stage,a new multiple models GGIW-CPHD based on best-fitting Gaussian approximation method (BFG)and strong track-ing filter (STF)was proposed.Firstly,on the basis of measurement set partitions,the GGIW-CPHD algorithm used the best-fitting Gaussian approximation method to implement the fusion of multiple models in the PHD predict stage.Then,a fading factor of strong tracking filter was used to correct the predict covariance matrix of the GGIW component.At last,the estimation of centroid kinematic state and extension state were completed in the CPHD update stage,and the probability of different tracking model was updated by the corresponding likelihood functions.The simulation results show that our algorithm can obtain the interacting multiple model performance under the GGIW-CPHD fil-ter framework,decrease the error of group targets state estimation in the maneuvering stage and process the group combination/spawning case effectively.关键词
群目标跟踪/最适高斯近似/强跟踪滤波/合并/衍生Key words
group targets tracking/best-fitting Gaussian approximation/strong tracking filter/com-bination/spawning分类
信息技术与安全科学引用本文复制引用
汪云,胡国平,甘林海..基于多模型GGIW-CPHD滤波的群目标跟踪算法[J].华中科技大学学报（自然科学版）,2017,42(2):89-94,6.基金项目
国家自然科学基金资助项目(61372166,61501495) （61372166,61501495）
