摘要
本文提出一种非线性量测数据融合方法,给出多传感器非线性量测数据融合公式与计算步骤,解决了在非线性量测数据融合时传统的极大似然方法和最小二乘方法需要求解非线性方程(组)的难题;同时提出一种非线性量测数据自校准融合方法,给出量测数据未知系统误差(未知输入)的自识别自校准公式与计算步骤,能够对非线性量测数据中事先无法校准的系统误差进行自动识别、估计、补偿和修正。文中详细讨论了秩采样量测数据融合方法和Sigma点采样量测数据融合方法,秩采样量测数据自校准融合方法和Sigma点采样量测数据自校准融合方法。大量实例计算和仿真模拟验证表明,本文方法不但比传统方法计算简单,便于工程应用,而且能够有效减小误差,显著提高精度。
A nonlinear measurement data fusion method is proposed.The multisensor nonlinear measurement data fusion formulas and calculation steps are given,which solve the problem that the traditional maximum-likelihood and least-squares methods need to solve nonlinear equations in nonlinear measurement data fusion.At the same time,a nonlinear measurement data self-calibration fusion method is proposed.The self-calibration formulas and calculation steps of the unknown system error(unknown input)in measurement data are given,which can automatically recognize,estimate,compensate and correct system errors that cannot be calibrated beforehand in nonlinear measurement data.In this paper,the rank sampling and Sigma-point sampling measurement data fusion methods as well as the rank sampling and Sigma-point sampling measurement data self-calibration fusion methods are discussed in detail.A large number of example calculations and simulation validations showthat the presented methods in this paper are not only simpler in calculation than traditional methods,but also more convenient for engineering applications,and can effectively reduce the errors and greatly improve the accuracy.
作者
傅惠民
崔轶
FU Huimin;CUI Yi(Research Center of Small Sample Technology,Beihang University,Beijing 100191,China)
出处
《智能计算机与应用》
2020年第7期5-10,共6页
Intelligent Computer and Applications
基金
国家重点基础研究发展计划(2012CB720000)
工信部2018年智能制造综合标准化项目《基于数字仿真的机械产品可靠性测试方法标准研究与试验验证》
关键词
非线性
量测
数据融合
自校准
多传感器
Nonlinear
Measurement
Data fusion
Self-calibration
Multisensor
