基于行人运动模态辨识的室内外无缝导航航向算法研究

发布时间：2018-02-07 10:27

本文关键词： 惯性传感器改进型互补滤波磁异常辨识自适应卡尔曼滤波行人运动模态辨识　出处：《传感技术学报》2017年06期 　论文类型：期刊论文

【摘要】：目前行人导航航向解算算法均基于导航传感器在行人身体上的固定安装模式,或者依赖其他射频信息辅助修正陀螺航向,这极大约束了导航传感器的适用条件。为此,利用陀螺对低频噪声的敏感性及加速度计低频的稳定性,提出了解决行人手持手机稳态查看与非稳态摇摆的运动模态辨识算法和基于时域互补滤波器实现姿态的最优融合方法;研究了改进型互补滤波以消除行人的运动加速度对姿态解算的干扰误差,提高了载体姿态的测量精度;此外,利用磁传感器标定后的数据设计了自适应卡尔曼滤波算法,抑制了航向角的误差发散。经实际数据测试验证,室内外行人手持稳态与非稳态下的航向角测量精度提高了80%,同时大大提高了导航传感器的适用性与便携性,满足实际工程的使用需求。
[Abstract]:At present, all of the navigation algorithms are based on the fixed installation mode of the navigation sensor on the pedestrian body, or rely on other radio frequency information to modify the gyroscope course, which greatly restricts the applicable conditions of the navigation sensor. Based on the sensitivity of gyroscope to low frequency noise and the stability of accelerometer at low frequency, a motion modal identification algorithm for steady viewing and unsteady swing of handheld mobile phone is proposed, and the optimal fusion method based on time domain complementary filter is proposed. The improved complementary filter is studied to eliminate the interference error of the pedestrian acceleration to the attitude calculation, and the measurement accuracy of the carrier attitude is improved. In addition, an adaptive Kalman filter algorithm is designed based on the calibrated data of the magnetic sensor. The error divergence of the heading angle is restrained. Through the actual data test, it is proved that the accuracy of the heading angle measurement under the steady and unsteady state of indoor pedestrian handheld has been improved by 80%, and the applicability and portability of the navigation sensor have been greatly improved. To meet the actual needs of the use of the project.
【作者单位】：南京航空航天大学自动化学院;
【基金】：国家自然科学基金项目(61533008,61673208,61533009,61374115) 江苏省六大人才高峰项目(2013-JY-013) 江苏高校优势学科建设工程项目和江苏省“物联网与控制技术”重点实验室基金项目中央高校基本科研业务费专项资金项目(NZ2016104,NS2017016) 江苏省“333工程”科研立项项目(BRA2016405) 留学人员择优项目(2016年) 航空科学基金项目(20165552043) 江苏省普通高校研究生科研创新计划项目(KYLX15_0264) 江苏省自然科学基金项目(BK20141453)
【分类号】：TN96;TP212

【参考文献】