Kalman滤波中测量粗差的探测与修复  

The Detection and Correction of Gross Error in Observed Values for the Kalman Filter

作　　者：宋迎春[1,2] 朱建军[1] 陈正阳[1] 郭云开[1] 

机构地区：[1]中南大学信息物理工程学院,长沙410083 [2]西安测绘研究所,西安710054

出　　处：《工程数学学报》2008年第3期437-441,共5页Chinese Journal of Engineering Mathematics

基　　金：国家自然科学基金(40574001,40774001);863专项基金(2007AA12Z331);湖南省自然科学基金(06JJ5131);湖南省教育厅规划课题(06C441).

摘　　要：在动态定位的数据处理中动态观测量可能存在粗差,若数据处理不考虑对这些异常的特别处理,则所提供的动态信息将极不可靠。文中计算了粗差对滤波结果的影响,并由此建立了动态定位中粗差的探测和修复算法。最后用数字仿真实验论证了方法的可行性。For the data process in kinematic positioning, there may be gross errors in dynamic observation values. If data processing does not pay attention to these gross errors, the information offered in kinematic positioning will not be extremely reliable. This paper calculates the influence of gross error on filter result, and gives an algorithm for detecting and correcting of gross errors in observed values for kinematic positioning. Lastly, a numerical simulation is carried out to prove the feasibility of this method.

关 键 词：卡尔曼滤波 动态定位 粗差 

分 类 号：O231.1[理学—运筹学与控制论]