...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of aerospace engineering >EM-FKF Approach to an Integrated Navigation System
【24h】

EM-FKF Approach to an Integrated Navigation System

机译：EM-FKF组合导航系统方法

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In this paper, to reduce the computational load of the federated Kalman filter, an expectation-maximization federated Kalman filtering (EM-FKF) algorithm for integrated navigation systems is proposed. First, the states with poor estimate accuracies are removed from local filters to reduce the computational load. Then the EM algorithm is applied. More precisely, the common states for each local filter are estimated in E-step of EM algorithm by using a linear Kalman filtering algorithm, and its own unique sensor biases are updated in M-step of EM algorithm. The M-step in the local filters and the fusion of common states in the master filter are performed simultaneously, so the computational burden is further reduced for the federated Kalman filter. The proposed algorithm was evaluated with simulated data first, then an experiment was conducted on a real inertial navigation system, global positioning system, and star sensor (INS/GPS/SS) integrated navigation system to verify the proposed algorithm. The results of the simulation and the experiment demonstrated that the proposed algorithm effectively reduced the computational load, compared with the standard federated Kalman filtering algorithm.
机译：为了减轻联合卡尔曼滤波器的计算量，提出了一种期望最大的联合导航系统联合卡尔曼滤波算法。首先，将估计精度不佳的状态从局部滤波器中删除，以减少计算量。然后应用EM算法。更精确地，使用线性卡尔曼滤波算法在EM算法的E步中估计每个局部滤波器的公共状态，并在EM算法的M步中更新其自身唯一的传感器偏置。局部滤波器中的M步和主滤波器中的公共状态融合是同时​​执行的，因此，联邦卡尔曼滤波器的计算负担进一步减轻。首先利用仿真数据对所提出的算法进行了评估，然后在真实惯性导航系统，全球定位系统和星际传感器（INS / GPS / SS）组合导航系统上进行了实验，以验证该算法。仿真和实验结果表明，与标准联合卡尔曼滤波算法相比，该算法有效地减少了计算量。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号