顾及空间分布和NLOS的UWB室内可信定位方法

针对现行室内可信定位存在的三维定位精度低与结果可信度难以定量评估的难点，提出了一种可信超宽带（UWB）室内定位模型。该方法采用无迹卡尔曼滤波和IGG-III抗差理论模型，解决了UWB定位中的非线性化和非视距误差问题；同时，为提高该方法定位结果的可信度，进一步构建了一种兼顾UWB基站空间分布和非视距误差的可信精度评估模型。与现有模型相比，提出的方案可有效实现基于UWB的室内高精度可信定位。仿真和实测数据结果表明：1）与传统最小二乘结果相比，该方法可提高30%~40%的UWB定位精度；2）引入IGG-III抗差模型可提高11%的定位精度；3）非线性化误差对UWB定位精度的影响不尽相同，约为22%~50%；4）提出的可信度评估模型可实时准确评估UWB定位精度。

An ultra-wideband based credible indoor positioning method by considering geometry-distribution and NLOS signal

In order to overcome the bottlenecks (i.e., low accuracy for 3D indoor positioning, low reliability, and no credible index) in indoor credible positioning at present, an ultra-wideband (UWB)-based credible indoor positioning method is presented. In this method, an unscented Kalman filter and an IGG-III robust algorithm are adopted together to overcome the problems of nonlinearization and non-line of sight errors. Meanwhile, a credible evaluation algorithm by considering the base station geometry-distribution and non-line of sight error is proposed to provide credible index for indoor positioning. Compared to the existing UWB positioning method, our method can provide users with high precision and credible indoor positioning solutions based on UWB. Results from both simulated data and experiment data illustrate that, 1) this method provides 30%~40% improvements on UWB positioning compared to the solutions calculated by traditional least square; 2) the UWB positioning accuracy can be furtherly improved by 11% by applying the IGG-III robust algorithm; 3) the impact of nonlinearization error on UWB positioning accuracy is varying with errors, changing from 22% to 50%; 4) it can accurately evaluate the accuracy of UWB positioning in real time.