摘要
针对矿山井下巡检机器人因受环境影响而定位精度低、实时性差等问题,提出了一种基于UWB(超宽带)与IMU(惯性测量单元)的融合定位方法。首先,利用UWB测距模块测量巡检机器人与UWB基站间的距离;其次,构建GRU(门控循环单元)神经网络模型,将UWB测量距离和真实距离输入神经网络模型中进行训练,得到GRU修正模型;然后,通过GRU神经网络修正模型对实测距离的修正,抑制复杂环境下NLOS(非视矩)距离误差,得到较之前更为精确的距离;最后,通过ESKF(误差状态卡尔曼滤波)将修正后的距离信息与IMU提供的数据结合,构建误差状态方程和量测方程,分别对系统的误差状态进行演化和更新,解算出更精确的位置坐标。试验结果表明:在三种不同NLOS环境下,修正后的静态实验定位精度较之前分别提升了6.09%、16.56%、18.89%,且在复合场景下定位精度较之前提升了11.36%;在动态实验中,修正后的平均定位精度较之前提升了11.10%。通过对非视距下测量的伪距进行修正可减小了非视距的影响,提高定位精度。
A fusion positioning method based on UWB(Ultra Wideband)and IMU(Inertial Measurement Unit)was proposed to address the issues of low positioning accuracy and poor real-time performance of underground inspection robots in mines due to environmental influences.Firstly,the UWBranging module was used to measure the distance between the inspection robot and the UWB base station.Secondly,a GRU(Gated Recurrent Unit)neural network model was constructed,and the UWB measurement distance and real distance were input into the neural network model for training,resulting in a GRU correction model.Then,the GRU neural network was used to correct the model for the measured distance,suppressingNLOS(Non-Line of Sight)distance errors in complex environments and obtaining more accurate distances than before.Finally,the corrected distance information was combined with the data provided by the IMU through ESKF(Error State Kalman Filter)to construct the error state equation and measurement equation,respectively,to evolve and update the system's error state and calculate more accurate position coordinates.The experimental results show that under three different NLOS environments,the corrected static experimental positioning accuracy is improved by 6.09%,16.56%and 18.89%respectively,and the positioning accuracy in the composite scene is improved by 11.36%compared to before.In the dynamic experiment,the corrected average positioning accuracy is improved by 11.10%compared to before.By correcting the pseudo range measured under non line of sight,the impact of non line of sight can be reduced and the positioning accuraey can be improved.
作者
蒯文博
张宏伟
蒲忠辉
肖卫
KUAI Wenbo;ZHANG Hongwei;PU Zhonghui;XIAO Wei(School of Electrical Engineering and Automation,Henan Polytechnic University,Jiaozuo,Henan 454003.China;Henan Key Laboratory of Intelligent Detection and Control of Coal Mine Equipment,Jiaozuo,Henan 454003.China;Henan Longyu Energy Co.,Ltd.,Shangqiu,Henan 476600.China)
出处
《矿业研究与开发》
北大核心
2025年第8期200-208,共9页
Mining Research and Development
基金
河南省高校科技创新团队项目(20IRTSTHN019)
