基于自适应PDR补偿的UWB/MIMU组合行人定位算法

UWB/MIMU Combined Pedestrian Location Algorithm Based on Adaptive PDR Compensation

为提高室内环境下行人的定位精度,针对超宽带(UWB)和微型惯性测量单元(MIMU)在行人室内定位中存在的定位不连续和定位误差累计问题,提出一种自适应行人航位推算(PDR)补偿的UWB/MIMU组合方案。首先提出一种多约束的PDR计步算法,降低了对阈值设置准确性的要求,消除了伪波峰和伪波谷对计步结果的影响;接着通过引入零速更新(ZUPT)和零角速率更新(ZARU)组合的方法对航向角误差进行修正;最后基于自适应滤波算法解算行人的位姿并进行了实验验证。结果表明,所提方案有效抑制了定位轨迹误差增大趋势,使得偏航角误差减少2°。

In order to improve pedestrian positioning accuracy in indoor environment,aiming at the problems of positioning discontinuity and positioning error accumulation existing in ultra wide band(UWB)and micro inertial measurement unit(MIMU)in pedestrian indoor positioning,an adaptive pedestrian dead reckoning(PDR)compensation combined UWB/MIMU scheme was proposed.Firstly,a multi-constraint PDR pedometer algorithm was proposed,which reduced the requirement for the accuracy of threshold setting and eliminated the influence of spurious peaks and troughs on the pedometer results.Then,the combination of zero velocity update(ZUPT)and zero angle rate update(ZARU)was introduced to correct the heading angle error.Finally,the adaptive filtering algorithm was used to solve the pose of the pedestrian and the experimental verification was carried out.The results show that the proposed scheme restrains the increasing trend of positioning trajectory error and reduces the yaw angle error by 2°.