基于室内无人车的UWB和IMU新型融合定位方法  

A New Positioning Method Combining of UWB and IMU Based on Indoor Unmanned Vehicles

作　　者：陈荆宇 徐琳[1] 黄斌[1] 李路航 Chen Jingyu;Xu Lin;Huang Bin

机构地区：[1]武汉理工大学汽车工程学院,湖北武汉430070

出　　处：《工程机械》2022年第11期1-7,I0009,共8页Construction Machinery and Equipment

基　　金：武汉理工大学国家大学生创新创业训练计划项目(3120400002360);湖北省重点研发计划项目(20201f0327-202007KJ01)。

摘　　要：|为提高室内无人车行驶精度和准确性,提出一种超宽带定位系统(Ultra-Wideband,UWB)与惯性导航系统(Inertial Navigation System,INS)数据融合的新型定位方法。该方法针对UWB定位中常见的非视距(Non Line of Sight,NLOS)误差,引入了惯性测量单元(Inertial Measure ment Unit,IMU),在常用的误差状态卡尔曼滤波的基础上,利用IMU的预测位置与UWB系统的伪距判断是否发生NLOS误差,通过迭代重加权最小二乘(Iterative Reweighted Least Squares,IRLS)处理UWB系统的伪距测量值,再由观测环节计算出最终位置,达到减少NLOS误差的目的。通过试验证实新型融合定位算法相较于传统的UWB和IMU数据融合定位方法降低了26.48%的最大NLOS误差。In order to improve the accuracy and precision of indoor unmanned vehicles, a new positioning method based on data combining of Ultra-Wideband(UWB) positioning system and inertial navigation system(INS) is proposed. The method, in view of the common non line of sight(NLOS) error in UWB positioning, introduces an inertial measurement unit(IMU), and based on the common error-state Kalman filter, uses the predicted position of the IMU and the pseudo-range of the UWB system to determine whether the NLOS error occurs. By iterative reweighted least squares(IRLS), it processes the pseudo-range measurements of the UWB system, and then calculates the final position from the observation link to achieve the purpose of reducing the NLOS error. It is experimentally confirmed that the proposed positioning algorithm reduces the maximum NLOS error by 26.48% compared to the conventional UWB and IMU data fusion positioning method.

关 键 词：室内无人车定位 超宽带 误差状态卡尔曼滤波 

分 类 号：TN925[电子电信—通信与信息系统] TN96[电子电信—信息与通信工程]