摘要
基于UWB技术的定位方法在现实生活中得到了广泛应用,但由于复杂的道路环境和障碍物影响,其产生的NLOS误差会严重影响系统的位置估计,导致定位精度低,鲁棒性差。为了提高复杂环境下UWB的定位方法的准确性和鲁棒性,提出了一种基于UWB和IMU数据融合的定位方法,该方法有效地将全局定位和局部定位结合起来,采用BP神经网络算法和EKF算法对IMU数据进行处理,并对IMU与UWB数据进行融合。实验结果表明,与传统UWB定位方法相比,该方法能有效抑制NLOS干扰对定位估计的影响,提高定位系统的精度和鲁棒性。
The positioning method based on UWB technology has been widely used in real life.Howerer,due to the influence of complex road conditions and obstacles,the NLOS error will seriously affect the position estimation of the system,resulting in low positioning accuracy and poor robustness.The IMU is a relatively simple way of positioning,but errors accumulate over time.In order to improve the accuracy and robustness of UWB based positioning method,this paper proposes a positioning method based on UWB and IMU data fusion.This method effectively combines global positioning and local positioning,and uses EKF algorithm and BP neural network oalgorithm to process IMU data and to fuse IMU and UWB data.Experimental results show that compared with the traditional ultra-wideband based localization method,this method can effectively suppress the influence of NLOS interference on localization estimation and improve the positioning accuracy and robustness of the system.
作者
王金柱
李骏驰
董亮
庞毅
梁茵
WANG Jin-zhu;LI Jun-chi;DONG Liang;PANG Yi;LIANG Yin(Tianjin Teda Binhai Clean Energy Group Co.,Ltd.,Tianjin 300300 China;Tianjin Chengjian University,Tianjin 300384 China)
出处
《自动化技术与应用》
2021年第4期19-23,共5页
Techniques of Automation and Applications
关键词
定位
扩展卡尔曼滤波
BP神经网络
多传感器融合
MATLAB仿真
positioning
Extended Kalman Filtering
Back Propagation Neural Network
multi-sensor fusion
MATLAB simulation
作者简介
王金柱(1984-),男,高级工程师,硕士,从事天然气工艺与相关自动化设备的技术研究工作。
