基于粒子滤波的PDR/地磁指纹室内定位

PDR/geomagnetic fingerprint indoor positioning method based on particle filter

针对行人航位推算(PDR)定位存在误差累积和地磁指纹不唯一导致的误匹配问题,本文改进了基于粒子滤波的PDR/地磁指纹室内定位方法。在PDR定位过程中利用地图信息控制粒子权重更新,得到较为准确的位置信息后,利用动态时间规整(DTW)算法在PDR推算位置基础上进行快速序列匹配,获取最优位置估计。试验结果表明,融合定位方法有效解决了行人位置穿墙问题,最大定位误差小于1.5 m,53.33%概率定位精度1 m。

Aiming at the problem of error accumulation in PDR positioning and mismatching caused by non-unique geomagnetic fingerprint,this paper combines PDR with geomagnetic fingerprint to conduct indoor positioning research,and proposes a PDR/geomagnetic fingerprint indoor positioning method based on particle filter. In the PDR positioning process,the map information is used to control the updating of particle weight to obtain more accurate position information. Then,dynamic time structuring algorithm is used to perform rapid sequence matching on the basis of PDR calculated position to obtain the optimal position estimation. The experimental results show that the fusion positioning method can effectively solve the problem of pedestrians’ position passing through the wall. The maximum positioning error is less than 1.5 m,and the positioning accuracy is 1 m with a 53.33% probability.