基于在线地磁指纹的航迹校准算法

设计一个行人航迹推断定位系统(PDR),该系统包括鞋载可穿戴MEMS惯性传感器模块、蓝牙通信模块、和用于执行校准算法的智能手机端.鞋载惯性导航模块利用位置估算、零速修正和步伐检测等算法获取消防人员位置估计,并通过低功耗蓝牙通信模块将消防员位置估计传到智能手机.在手机端,利用智能手机自带地磁传感器及加速度传感器提取姿态无关的三维地磁指纹,根据姿态无关三维地磁指纹生成正、反向地磁轨迹,将消防员位置估计与地磁轨迹相关联,再通过DTW子序列算法实现地磁轨迹精准匹配,实现消防员航迹校准,从而有效解决惯性导航中累计误差无法消除的难题,最终实现消防员室内复杂场景精确定位.大量测试实验表明,该系统在办公室和超市两种典型应急消防救援场景分别可获得80％绝对误差1.4～2m以内.

Online magnetic fingerprints based personal dead reckoning calibration algorithm

A personal dead reckoning (PDR) localization system is designed in this paper.PDR system include a foot-mounted wearable MEMS inertial sensors module and a smartphone client,which used to run a calibration algorithm.The foot-mounted inertial module obtains rescuer`s position estimation by using position estimation,zero velocity update and step detection algorithms.The foot-mounted inertial module transfers rescuer`s position estimation to the smartphone over the Bluetooth low energy wireless interface.In the smartphone client,extracting the attitude-free magnetic fingerprint using the acceleration and gyroscope sensor integration with smartphone,generating forward and backward magnetic trajectory according to the attitude-free magnetic fingerprint,all the attitude-free magnetic fingerprints are configured with the location of PDR.The magnetic trajectory matching based PDR location calibration algorithm which leverages DTW subsequence can effectively resolve the problem of the PDR suffers from the severe position error accumulation with time due to the inaccurate step length and moving direction estimation,accordingly realizing the purpose of accurate indoor localization for rescuers.The vast experimental be carried out in two representative indoor environments (office and supermarket) shows that the proposed the proposed self-localization algorithm can obtain an 80 percentile localization accuracy of 1.4 m and 2 m.