复杂环境下的UWB/PDR紧组合定位方法

Tight integration positioning method of UWB/PDR in complex environments

针对当前室内定位需求呈爆发式增长的趋势,基于超宽带(UWB)的测距信息和惯性传感器信息,提出了一种改进的UWB与行人航位推算(PDR)紧组合定位方法。与松组合不同,UWB/PDR紧组合直接利用UWB测距值进行融合,不须单独解算UWB位置信息;基于卡方(χ^(2))检验识别UWB非视距(NLOS)误差,并构造等价权因子自适应调节量测噪声协方差矩阵,降低含有粗差或NLOS误差的UWB测距的权重。不仅抑制了PDR漂移误差,而且在UWB存在NLOS误差和基站信号缺失等复杂环境下仍能正常工作,增加了系统的鲁棒性。实验结果表明:在UWB NLOS环境下,相较松组合将NLOS误差引入UWB定位结果,紧组合基于χ~2检验可以有效削弱NLOS误差的影响,定位精度明显优于松组合;而在UWB基站信号缺失的环境下,因松组合的定位有可能退化为PDR定位结果,其误差迅速漂移,UWB/PDR紧组合更加充分利用了UWB测距信息,仍能保持较好的定位效果,更适合复杂环境的位置估计。

In response to the current trend of explosive growth of indoor positioning demand, this paper proposes an improved ultra wide band(UWB) and pedestrian dead reckoning(PDR) tight integration positioning method based on the range information of UWB and inertial sensor information. The tight integration directly uses the UWB range values for fusion without solving the UWB position information separately, identifies the UWB non line of sigh(NLOS) error based on the χ^(2) test.It not only suppresses the PDR drift error, but also increases the robustness of the system in complex environments. The experimental test shows that in the UWB NLOS environment, compared with the loose combination, the NLOS error is introduced into the UWB positioning result,the tight integration can effectively weaken the influence of NLOS error based on the χ~2 test, and the localization accuracy is significantly better. In the environment where the UWB base station signal is missing, the UWB/PDR tight integration makes more full use of the UWB ranging information, still maintains a good positioning effect, and is more suitable for position estimation in complex environments.