首页 | 本学科首页   官方微博 | 高级检索  
     

基于系统误差和状态联合估计的目标跟踪算法
引用本文:胡玉梅,胡振涛,郑珊珊,李贤,郭振.基于系统误差和状态联合估计的目标跟踪算法[J].计算机科学,2015,42(11):310-313.
作者姓名:胡玉梅  胡振涛  郑珊珊  李贤  郭振
作者单位:河南大学图像处理与模式识别研究所 开封475004,河南大学图像处理与模式识别研究所 开封475004,河南大学图像处理与模式识别研究所 开封475004,河南大学图像处理与模式识别研究所 开封475004,河南大学图像处理与模式识别研究所 开封475004
基金项目:本文受国家自然科学基金项目(61300214,U1204611,3),河南省高校科技创新团队支持计划(13IRTSTHN021),中国博士后基金(2014M551999),河南省青年骨干教师资助
摘    要:针对线性系统中系统误差对状态估计精度造成的不利影响,在卡尔曼滤波算法框架下提出一种基于系统误差和状态联合估计的目标跟踪算法。在算法实现过程中,首先定量分析了系统误差对目标状态估计及其估计误差协方差矩阵的影响,进而结合状态扩维技术构建系统误差配准的实现过程,最终依据标准卡尔曼滤波迭代流程设计了算法实现步骤。仿真实验结果表明: 在系统误差恒定和时变两种情况下,新算法在系统误差配准和状态估计上具有可行性和有效性。

关 键 词:状态估计  系统误差配准  目标跟踪  卡尔曼滤波
收稿时间:2014/11/4 0:00:00
修稿时间:2015/3/20 0:00:00

Novel Target Tracking Algorithm Based on Joint Estimation of System Error and State
HU Yu-mei,HU Zhen-tao,ZHENG Shan-shan,LI Xian and GUO Zhen.Novel Target Tracking Algorithm Based on Joint Estimation of System Error and State[J].Computer Science,2015,42(11):310-313.
Authors:HU Yu-mei  HU Zhen-tao  ZHENG Shan-shan  LI Xian and GUO Zhen
Affiliation:Institute of Image Processing and Pattern Recognition,Henan University,Kaifeng 475004,China,Institute of Image Processing and Pattern Recognition,Henan University,Kaifeng 475004,China,Institute of Image Processing and Pattern Recognition,Henan University,Kaifeng 475004,China,Institute of Image Processing and Pattern Recognition,Henan University,Kaifeng 475004,China and Institute of Image Processing and Pattern Recognition,Henan University,Kaifeng 475004,China
Abstract:Aiming at adverse effects resulted from system error on state estimation precision in linear system,a novel target tracking algorithm based on joint estimation of system error and state was proposed in Kalman filter framework.Firstly,the influence of system error on target state estimation and state estimation error covariance matrix were analyzed quantitatively.Secondly,combined with the extension method of state dimensions,the registration process of system error was constructed.Finally,the realization steps of new algorithm were designed according to the iterative process of standard Kalman filter.Simulation results show the feasibility and effectiveness of new algorithm dealing with system error registration and state estimation when system error is constant or time-vary.
Keywords:State estimation  System error registration  Target tracking  Kalman filter
本文献已被 万方数据 等数据库收录!
点击此处可从《计算机科学》浏览原始摘要信息
点击此处可从《计算机科学》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号