基于Stirling插值滤波和扩展卡尔曼滤波的方位估计(英文)  

Stirling’s Interpolation Filtering and Extended Kalman Filtering for Orientation Estimation

作　　者：伍雪冬[1] 蒋新华[1] 宋执环[2] 谭文[3] 

机构地区：[1]福建工程学院电子信息与电气工程系,福州350014 [2]浙江大学工业控制技术国家重点实验室,杭州310027 [3]湖南科技大学信息与电气工程学院,湘潭411201

出　　处：《系统仿真学报》2007年第19期4464-4468,共5页Journal of System Simulation

基　　金：The project was supported by the Natural Science Foundation of Fujian province of China (2006J0017);the Youth Innovation Project of Science and Technology of Fujian province (2005J048);The education department’s Science and Technology planned project of Fujian province (JA05308;JB05287).

摘　　要：插值间距的适当选择可以从某种程度上确保剩余项更接近于全Taylor级数的高阶项,因此,对大多非线性比较强的状态估计问题来说,基于Stirling插值原理的SIF要比EKF具有更大的优越性。估计方位时我们用仿真实验分别对SIF和EKF的特性进行了比较,用四元数而不用欧拉角来表示旋转量以消除方位估计的奇异性问题,状态向量包括四元数方位和旋转速度,以四元数作为测量输入,因此测量方程是线性,仿真实验结果表明:由于动态系统的准线性本质,估计四元数方位时EKF特性要比SIF好。A sensible choice of internal length can ensure that the remainder terms in some sense will be close to the higher-order terms of full Taylor series, so the Stirling＇s interpolation filtering （SIF） based on Stirling＇s interpolation formula is a superior alternative to the extended Kalman filter （EKF） for a variety of highly nonlinear state estimation problems. The simulation experimental study was presented to compare the performance of SIF and EKF for orientation estimation. Using quaternion rather than Euler angles to represent rotation, this eliminates the singularity problem of orientation denotation. The state vector includes quaternion orientation and rotation velocity, and the quaternion is used as measurements. As a result of this approach, the measurement equations become linear. Simulation results of the SIF and EKF with synthetic data prove EKF is a better choice for quaternion orientation estimation due to the quasi-linear nature of the dynamics system.

关 键 词：方位估计 斯梯林插值滤波 扩展卡尔曼滤波 四元数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]