基于反馈校正的WLAN与PDR融合定位方法研究*

在智能手机作为终端的室内定位系统中,基于WLAN的指纹定位和行人航位推算定位(PDR)是较为常用的方法。WLAN定位方式能够进行独立定位,但稳定性差;PDR定位短时间内稳定性好,但易产生累积误差且不能进行独立定位。针对上述问题,本文通过分析两种定位方式的误差特点,提出一种基于反馈校正的融合定位方法。该方法主要分为两个过程:基于自适应粒子分布的信息融合过程和基于融合信息的PDR自适应线性反馈校正过程。利用提出的融合方法,可以很好地解决一般融合方法所存在的定位结果稳定性差的问题。实验数据表明,本文提出的融合定位方法对最终的定位结果有较大的改善效果。

Research on the feedback correction-based fusing method for WLAN and PDR positioning

In the indoor positioning systems using smart phone as the terminal, WLAN fingerprint positioning and Pedestrian dead reckoning (PDR) are two common ways. WLAN fingerprint positioning can work independently, but the stability is very poor. The positioning accuracy is high for PDR in a short time, however it can''t independently give positioning result and has serious cumulative error. To address this issue, a Feedback correction based fusing method is presented by analyzing the error characteristics of these two ways for positioning. The scheme consists of two processes. The first is integration process which based on adaptive particle distribution, and the second is the adaptive linear feedback correction process. With the proposed method, the problem of poor stability in general fusing methods can be solved. Experimental results show that the proposed fusing method can well improve the stability and accuracy of the final positioning result.