PDR算法对地磁室内定位精度的提升研究

利用建筑物中金属结构引起的地磁场扰动可以对室内的行人目标进行定位，而且基于地磁场的定位无需布设任何额外设施，因此可以以低成本实现定位。但仅靠单一的地磁技术无法满足室内定位的精度要求。为了解决磁场数据中单点定位的模糊性问题，本文提出了一种利用粒子滤波算法将PDR与地磁相融合的室内定位方法，并开发了地磁室内导航系统，以智能手机为硬件平台构建磁力计传感器模型，建立匹配轨迹的均方误差准则并实现PDR累积误差实时校正的迭代计算。在68 m×1.8 m的试验区域内，产生的平均定位误差为1.13 m，最大定位误差为2.17 m。本文算法的定位精度比单独PDR算法提升了42%；与单一地磁指纹匹配算法相比，定位精度提高了57%。试验证明，本文提出的融合算法对提高室内定位精度具有显著的作用。

PDR algorithm improves indoor positioning accuracy of geomagnetism

The use of geomagnetic field disturbances caused by metal structures in the building can locate indoor pedestrian targets, and the positioning based on the geomagnetic field does not require any additional facilities, so positioning can be achieved at low cost. However, only a single geomagnetic technique could not meet the accuracy requirements of indoor positioning. In order to solve the ambiguity problem of single point positioning in magnetic field data, we propose an indoor positioning method that uses particle filter algorithm to fuse PDR with geomagnetism and develops. The geomagnetic indoor navigation system uses the smart phone as the hardware platform to construct the magnetometer sensor model, establishes the mean square error criterion of the matching trajectory and realizes the iterative calculation of the PDR cumulative error real-time correction. In the experimental area of 68 m×1.8 m, the average positioning error is 1.13 m and the maximum positioning error is 2.17 m. The positioning accuracy of the fusion algorithm proposed in this paper is 42% higher than that of the single PDR algorithm. Compared with the positioning of a single geomagnetic fingerprint matching algorithm, the positioning accuracy is improved by 57%. Experiments show that the proposed fusion algorithm has a significant effect on improving indoor positioning accuracy.