摘要
当前智能手机大多集成了多种传感器,组合多种传感器数据进行室内定位已成为研究热点。本文针对传统的Wi-Fi位置指纹室内定位精度不高的问题,提出了一种基于卡尔曼滤波结合行人航位推算的融合定位方法,经测试,融合定位方法的平均定位精度为3.1 m,较单独Wi-Fi指纹定位精度提升了43%。同时针对PDR步数检测不准确的问题,本文提出了基于低通滤波和阈值滤波的组合滤波方法,将计步准确率提高至99%,对于不同的应用场景具有较强的适应性。
At present,most smartphones integrate multiple sensors and combining multiple sensor data for indoor positioning has become a research hotspot.Aiming at the problem of low indoor positioning accuracy of traditional Wi-Fi location fingerprint,this paper proposes a fusion positioning method based on Kalman filter and pedestrian dead reckoning.Through testing,the average positioning accuracy of fusion positioning method is 3.1 m,which is 43%higher than that of Wi-Fi fingerprint alone.At the same time,in order to solve the problem of inaccurate PDR step detection,this paper proposes a combination filtering method based on low-pass filtering and threshold filtering,which improves the step counting accuracy to 99%,and has strong adaptability to different application scenarios.
作者
关沧海
王明亮
赵忠海
GUAN Canghai;WANG Mingliang;ZHAO Zhonghai(The Second Geodetic Survey Team of MNR,Harbin 150025,China)
出处
《测绘与空间地理信息》
2023年第S01期262-265,272,共5页
Geomatics & Spatial Information Technology
关键词
WI-FI
行人航位推算
组合滤波
卡尔曼滤波
融合定位
Wi-Fi
pedestrian dead reckoning(PDR)
combined filtering
kalman filter
fusion positioning
