Sui Xin, Song Jiapeng, Wang Changqiang, Ding Wenhao, Gao Song, Shi Zhengxu
School of Geomatics, Liaoning Technical University, Fuxin, 123000, China.
Dai Shi Intelligent Technology Co, Shanghai, 201802, China.
Sci Rep. 2024 Aug 27;14(1):19879. doi: 10.1038/s41598-024-70802-y.
A dynamic nonline-of-sight (NLOS) angle discrimination method is proposed to address the insufficiency of current research on the NLOS error characteristics of ultrawideband (UWB) signals in dynamic environments as well as the problem that UWB signals frequently suffer from occlusion, leading to poor or impossible localization. The experimental results indicate that the degree of UWB signal occlusion increases as the horizontal angle decreases, and when the horizontal angle is less than 167°, the UWB ranging error is so large that no ranging value is available. On this basis, a tightly integrated UWB/MEMS IMU positioning algorithm incorporating NLOS angle discrimination and map constraints is proposed; it employs horizontal angles to discriminate UWB ranging values in NLOS environments in accordance with the dynamic NLOS characteristics of UWB signals to assign better weights to UWB observations. Through comparative analysis of the results from both groups of experiments, the algorithm achieved northward, eastward, and planar positioning errors of 0.189 m and 0.126 m, 0.119 m and 0.134 m, 0.243 m and 0.211 m, respectively. Compared to the Robust Adaptive Kalman Filtering algorithm, the positional accuracy in the plane improved by 22.9% and 28.5%, respectively.
提出了一种动态非视距（NLOS）角度判别方法，以解决当前对动态环境中超宽带（UWB）信号的NLOS误差特性研究不足的问题，以及UWB信号经常遭受遮挡导致定位效果差或无法定位的问题。实验结果表明，UWB信号的遮挡程度随着水平角度的减小而增加，当水平角度小于167°时，UWB测距误差过大以至于没有可用的测距值。在此基础上，提出了一种结合NLOS角度判别和地图约束的紧密集成UWB/MEMS IMU定位算法；它根据UWB信号的动态NLOS特性，利用水平角度在NLOS环境中判别UWB测距值，以便为UWB观测值赋予更好的权重。通过对两组实验结果的对比分析，该算法在北向、东向和平面定位误差分别为0.189 m和0.126 m、0.119 m和0.134 m、0.243 m和0.211 m。与鲁棒自适应卡尔曼滤波算法相比，平面定位精度分别提高了22.9%和28.5%。
