| 引用本文: | 刘利军， 涂国勇， 朱时银， 李 曦. 外测数据误差的高精度估计与分离方法[J]. 雷达科学与技术, 2020, 18(6): 605-610.[点击复制] |
| LIU Lijun, TU Guoyong, ZHU Shiyin, LI Xi. High Precision Estimation and Separation Method of External Measurement Data Error[J]. Radar Science and Technology, 2020, 18(6): 605-610.[点击复制] |
|
| 摘要: |
| 航天发射场火箭外弹道数据预处理主要完成原始测量数据随机误差和系统误差的修正，是后续进行航迹融合和目标识别的基础。为解决传统最小二乘方法难以对测量数据随机误差进行高精度估计的难题，本文提出利用小波变换的多层分解与重构技术对测量数据的随机误差进行统计和消除；并针对复杂模型下测量数据系统误差难以分离的难题，采用模型辨识和回归分析理论对系统误差进行模型分析和诊断的方法。仿真和实际应用证明新算法简单适用，可有效解决传统方法难于对测元误差进行准确修正的问题，大大提高参与融合求解的测元数据质量和弹道解算精度，在靶场外测数据预处理方面具有良好的应用前景。 |
| 关键词: 外测 随机误差 系统误差 小波 回归分析 |
| DOI：DOI:10.3969/j.issn.1672-2337.2020.06.005 |
| 分类号:TN959.6；V557 |
| 基金项目: |
|
| High Precision Estimation and Separation Method of External Measurement Data Error |
|
LIU Lijun, TU Guoyong, ZHU Shiyin, LI Xi
|
|
Unit 63620 of PLA，Jiuquan 732750,China
|
| Abstract: |
| Pre-processing of rocket trajectory data in space launch site mainly completes the correction of random error and systematic error of original measurement data, which is the basis of track fusion and target re-cognition. In order to solve the problem that the traditional least square method is difficult to estimate the random error of measurement data with high precision, this paper proposes to use the multi-layer decomposition and reconstruction technology of wavelet transform to count and eliminate the random error of measurement data. Aiming at the difficult problem of separation of measurement data system error under complex model, the model 〖KG-*4〗identification and regression analysis theory are used to analyze and diagnose the system error. Simulation and practical application prove that the new algorithm is simple and applicable. It can effectively solve the problem that the traditional method is difficult to correct the error of the measuring element accurately, greatly improve the quality of the measuring element and the accuracy of trajectory calculation, and has a good application prospect in the pre-processing of the data measured outside the shooting range. |
| Key words: external measurement random error system error wavelet regression analysis |
