| 提高惯导系统Unscented卡尔曼滤波实时性技术 |
| |
| 引用本文： | 周战馨,陈家斌.提高惯导系统Unscented卡尔曼滤波实时性技术[J].火力与指挥控制,2007,32(6):37-39. |
| |
| 作者姓名： | 周战馨 陈家斌 |
| |
| 作者单位： | 1. 同济大学电子与信息工程学院,上海,200092
2. 北京理工大学,北京,100081 |
| |
| 摘 要： | Unscented卡尔曼滤波(UKF)通过构造一组,具有与给定状态估计相同的一、二阶矩,也可能是高阶矩的样本点,实现对非线性系统的状态估计,因此其计算效率取决于能够捕获这些数字特征的样本点数目.通过构造n 2个样本点来捕获n维状态变量的均值和方差,提出了将这种方法应用于惯性导航的初始对准.仿真结果表明在同等滤波精度的情况下,该方法比UKF计算效率更高,实时性更好.
|
| 关 键 词： | 非线性滤波 Unscented卡尔曼滤波 初始对准 惯导系统 Unscented Kalman Filter 卡尔曼滤波 实时性技术 Alignment Improvement 情况 精度 仿真结果 初始对准 惯性导航 应用 方法 方差 均值 状态变量 数字特征 计算效率 非线性系统 样本点 |
| 文章编号： | 1002-0640(2007)06-0037-03 |
| 修稿时间： | 2005年10月15 |
Research on the Improvement of Real-time Unscented Kalman Filter in INS Alignment |
| |
| ZHOU Zhan-xin,CHEN Jia-bin.Research on the Improvement of Real-time Unscented Kalman Filter in INS Alignment[J].Fire Control & Command Control,2007,32(6):37-39. |
| |
| Authors: | ZHOU Zhan-xin CHEN Jia-bin |
| |
| Abstract: | Unscented Kalman filter state estimates of the nonlinear system by constructing a set of sample points,which has the same known first and second moments,may be higher moments with the state.Its computational efficiency depends on the number of the sample points.A set of n+2 sample points captures first and second moments of n-dimensional state.This paper presents the application of this method in INS alignment.The simulation result shows that its has higher computational efficiency than unscented Kalman filter,under the second moment filter precision. |
| |
| Keywords: | nonlinear filter Unscented Kalman filter initial alignment |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
