基于行人运动模态辨识的室内外无缝导航航向算法研究

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.

Research on Indoor and Outdoor Seamless Navigation Heading Algorithm Based on Pedestrian Modal Identification

Nowadays,heading algorithm in pedestrian navigation is mostly based on the fixed mode of navigation sen-sor on the pedestrian body,or relies on other auxiliary radio frequency information,which greatly reduces the porta-bility of sensors. Hence,based on the sensitivity of the gyroscope to the low frequency noise and the stability of the accelerometer,this paper puts forward one solution to determine the steady-state and non-steady state of the hand held mobile based on the modal identification and proposes complementary filter to achieve optimal attitude based on time domain. This paper uses modified complementary filter to weaken the interference caused by pedestrian movement and improves the accuracy of the carrier attitude measurement. In addition,the adaptive kalman filter al-gorithm is designed by using the magnetic sensor calibration data to restrain the divergence of heading angle error. The test results show that this algorithm can ensure that the accuracy of heading angle measurement in indoor and outdoor pedestrians can be improved by 80% and greatly improve the adaptability and portability of sensors simulta-neously,which meets the demand of practical engineering.