|
|||||
|
|
| 基于统计学习的运动目标轨迹跟踪 | |
| 引用本文： | 束锋,李兴伟,王东明,汪茂,陆锦辉,朱伟强.基于统计学习的运动目标轨迹跟踪[J].高技术通讯,2009,19(12). |
| 作者姓名： | 束锋 李兴伟 王东明 汪茂 陆锦辉 朱伟强 |
| 作者单位： | 1. 南京理工大学电光学院,南京,210094 2. 东南大学移动通信国家重点实验室,南京,210092 3. 中国航天科工集团第8511研究所,南京,210094 |
| 基金项目： | 国家自然科学基金，863计划，中国航天科工集团第8511研究所航天基金 |
| 摘 要： | 为了更准确地跟踪被动辐射源的运动轨迹,提出了一种将到达时差(TDOA)最大似然定位方法与基于统计学习的算法相结合的轨迹跟踪算法.该算法利用TDOA最大似然网格搜索法对目标进行多次单独定位,获得目标初始轨迹,然后应用统计学习算法对轨道进行学习和拟合,输出运动轨迹.仿真表明,同卡尔曼滤波器相比,该方法提供了更高的跟踪精度,当拟合点数达到300个点时,最终的跟踪精度同单次定位精度相比可提高一个数量级. |
| 关 键 词： | 到达时差(TDOA) 运动目标 跟踪精度 卡尔曼滤波器 统计学习 |
Moving target locus tracking based on statistic learning |
|
| Shu Feng,Li Xingwei,Wang Dongming,Wang Mao,Lu Jinhui,Zhu Weiqiang.Moving target locus tracking based on statistic learning[J].High Technology Letters,2009,19(12). | |
| Authors: | Shu Feng Li Xingwei Wang Dongming Wang Mao Lu Jinhui Zhu Weiqiang |
| Affiliation: | Shu Feng,Li Xingwei,Wang Dongming~*,Wang Mao,Lu Jinhui,Zhu Weiqiang~(**) (School of Electronic Engineering and Optoelectronic Technique,Nanjing University of Science and Technology,Nanjing 210094) ( * National Mobile Communication Research Laboratory,School of Information Science and Engineering,Southeast University,Nanjing 210092) (** 8511 Institution,China Aerospace Science & Industry Corporation,Nanjing 210094) |
| Abstract: | To track the locus of a moving target more precisely, the paper proposes a target tracking method combining the statistic learning algorithm with the maximum likelihood (ML) location of time difference of arrival (TDOA). This algorithm uses the ML grid search method based on TDOA to obtain the initial locus, and then adopts statistical learning to fit this locus and output the final moving turget locus. The simulations show that this method provides a better tracking precision than the Kalman filter; when t... |
| Keywords: | time difference of arrival (TDOA) moving target tracking precision Kalman filter statistic learning |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |