一种多元信息融合的室内定位方法

发布时间：2018-04-11 20:45

本文选题：室内定位 + WiFi　；参考：《微型机与应用》2016年22期

【摘要】：针对现有室内定位技术在定位精度方面的不足,基于智能手机获取的多元传感器信息,提出一种多元信息融合的室内定位方法。首先,利用基于KNN(K-Nearest Neighbor)的Wi Fi指纹滤波法得到行人初始位置坐标,通过行人航迹推算技术估算下一个时刻位置;然后利用WiFi和地磁加权平均后的定位坐标对行人运动轨迹进行修正;最后对获取的位置信息进行卡尔曼滤波后得到了行人的最终位置。此外,设计了基于Android平台的手机室内定位软件,实现了高精度的轨迹生成。实验结果表明,该方法有效提高了定位的精确度,平均定位误差降至1 m。
[Abstract]:In view of the shortcomings of the existing indoor positioning technology in positioning accuracy, a multi-information fusion indoor location method is proposed based on the multi-sensor information obtained by smart phone.Firstly, the initial position coordinates of pedestrians are obtained by using the WiFi fingerprint filtering method based on KNN(K-Nearest neighbor, and the next time position is estimated by the footpath estimation technique.Then the pedestrian trajectory is corrected by WiFi and the location coordinates after geomagnetic weighted average. Finally, the final position of the pedestrian is obtained by Kalman filtering.In addition, the mobile phone indoor positioning software based on Android platform is designed to generate high precision trajectory.The experimental results show that this method can effectively improve the accuracy of positioning, and the average positioning error is reduced to 1 m.
【作者单位】：解放军理工大学国防工程学院;海军海防工程研究中心;清华大学建筑信息模型联合研究中心;
【分类号】：P228

【相似文献】