南京理工大学学报（自然科学版）2016，Vol.40Issue(1)：56-60,66,6.DOI:10.14177/j.cnki.32-1397n.2016.40.01.009
汽车毫米波雷达目标跟踪的快速平方根CKF算法
Fast square root CKF for automotive millimeter-wave radar target tracking
摘要
Abstract
QR decomposition of error covariance matrix of the square root cubature Kalman Filter ( CKF) is directly used for filter's prediction and updating,which can avoid breaking down when the covariance matrix is non-positive-definite. However, the prediction and updating process based on sigma-point propagation leads to obviously high computational burden. In this paper, a fast square root CKF method for automotive millimeter-wave radar target tracking is proposed. In this method,the Kalman equations are used to predict the state means and covariance during the prediction process, while square root CKF equations are used to compute the Kalman gain and update the state means and covariance during the updating process. Many experimental results show,either on efficiency or precision,our proposed method is superior to the similar square root unscented Kalman filter( UKF) and square root CKF algorithms.关键词
毫米波雷达/目标跟踪/平方根容积卡尔曼滤波器Key words
millimeter-wave radar/target tracking/square root cubature Kalman filter分类
信息技术与安全科学引用本文复制引用
刘华军,赖少发..汽车毫米波雷达目标跟踪的快速平方根CKF算法[J].南京理工大学学报（自然科学版）,2016,40(1):56-60,66,6.基金项目
国家"863"高技术研究计划资助项目(2015AA8106043) （2015AA8106043）
国家自然科学基金(61402237,61302156) （61402237,61302156）
