| 206 | 2 | 108 |
| 下载次数 | 被引频次 | 阅读次数 |
针对最小二乘法(Least Squares Method, LSM)对于超宽带(Ultra Wide Band, UWB)中飞行时间(Time of Flight, TOF)法在非视距(Non Line of Sight, NLoS)下定位精度低下的问题，提出了基于扩展灰狼算法(Extended Gray Wolf Algorithm, EGWO)优化后的长短期记忆(Long Short-Term Memory, LSTM)网络改进LSM定位算法(EGWO-LSTM-LSM)。采用LSTM及改进的EGWO建立最优测距误差预测模型，根据预测结果构造权重矩阵，在LSM上加权计算，并添加测距误差校正项，以改进LSM实现静态定位，并结合卡尔曼滤波器(Kalman Filter, KF)实现动态定位追踪。仿真结果表明，EGWO-LSTM预测准确率达98.857%,EGWO-LSTM-LSM将二维和三维位置误差分别稳定控制在10～25 mm,进一步提升了TOF定位精度。
Abstract：To solve the problem of low localization accuracy of Time of Flight(TOF) in Ultra Wide Band(UWB) under Non Line of Sight(NLoS) by Least Squares Method(LSM), an Extended Gray Wolf Algorithm(EGWO) is proposed to optimize the Long Short-Term Memory(LSTM) network that is then used to improve the LSM localization algorithm(EGWO-LSTM-LSM). First, LSTM and EGWO are used to establish the optimal range error prediction model, then the weight matrix is constructed according to the prediction results, weighted calculation is made on the LSM, and the range error correction item is added. Finally, the improved LSM is used to achieve static positioning, and the Kalman Filter(KF) is combined to achieve dynamic positioning and tracking. The simulation results show that the prediction accuracy of EGWO-LSTM is 98.857%, and the two-dimensional and three-dimensional position errors by EGWO-LSTM-LSM are stably controlled within 10～25 mm respectively, which further improve the positioning accuracy of TOF.
[1] HAN Y X,ZHANG X M,LAI Z X,et al.TOF-based Fast Self-positioning Algorithm for UWB Mobile Base Stations[J].Sensors,2021,21(19):6359.
[2] ALARIFI A,AL-SALMAN A,ALSALEH M,et al.Ultra Wideband Indoor Positioning Technologies:Analysis and Recent Advances[J].Sensors,2016,16(5):707.
[3] MAZHAR F,KHAN G M,S?LLBERG B.Precise Indoor Positioning Using UWB:A Review of Methods,Algorithms and Implementations[J].Wireless Personal Communications,2017,97(3):4467-4491.
[4] 徐建华，张雨霖，韩勇强.基于移动节点辅助定位的UWB室内定位方法[J].中国惯性技术学报，2023,31(2):141-147.
[5] JUAN C W,HU J S.Single-object Localization Using Multiple Ultrasonic Sensors and Constrained Weighted Least-squares Method[J].Asian Journal of Control,2021,23(3):1171-1184.
[6] RETSCHER G,KISS D,GABELA J.Fusion of GNSS Pseudoranges with UWB Ranges Based on Clustering and Weighted Least Squares[J].Sensors,2023,23(6):3303.
[7] TONG Z X,XUE J H,KANG Z Q.A Novel Indoor Positioning Algorithm Based on UWB[J].International Journal of Sensor Networks,2022,40(4):238-249.
[8] LIU A,LIN S W,WANG J G,et al.A Succinct Method for Non-Line-of-Sight Mitigation for Ultra-wideband Indoor Positioning System[J].Sensors,2022,22(21):8247.
[9] 秦明峰，胡丽格.一种基于伪距残差加权的最小二乘定位算法[J].计算机仿真，2022,39(12):44-48.
[10] 胡仲勋，杨旭静，金湘中.LSM算法评定空间直线度误差的分析与改进[J].湖南大学学报(自然科学版),2010,37(2):27-31.
[11] 刘公绪，李龙，路建民，等.无线信号交汇定位的奇异解问题表征与解决[J].中国科学：信息科学，2022,52(11):1992-2010.
[12] 肖岚，温丽丽，赵静，等.采用改进多目标灰狼算法的电力系统调度规划[J].机械设计与制造，2024(2):373-379.
[13] 马高权，周娜，谢蒙飞，等.基于GWO-KELM模型的变压器油纸套管典型绝缘故障辨识方法[J].电网与清洁能源，2023,39(5):38-48.
[14] GU J,JIAO L C,YANG S Y,et al.Fuzzy Double C-means Clustering Based on Sparse Self-representation[J].IEEE Transactions on Fuzzy Systems,2018,26(2):612-626.
[15] GAO D Q,ZENG X Y,WANG J Y,et al.Application of LSTM Network to Improve Indoor Positioning Accuracy[J].Sensors,2020,20(20):5284.
[16] 冯铃，张楚，刘伟渭.基于MGWO-SCN的滚动轴承故障诊断方法[J].机电工程，2022,39(10):1382-1389.
[17] 王述红，魏崴，韩文帅，等.基于CGWO算法的边坡最小安全系数全局寻优方法[J].东北大学学报(自然科学版),2022,43(7):1033-1042.
[18] 张涛，王亚刚，李开言.PCR仪器的IGWO-BP神经网络PID控制[J].控制工程，2023,30(5):822-829.
[19] 杨秀建，皇甫尚昆，颜绍祥.基于改进UKF的UWB/IMU/里程计融合定位方法[J].中国惯性技术学报，2023,31(5):462-471.
[20] 孙伟，孙沛伦.非视距环境下的UWB/INS室内定位方法[J].测绘科学，2023,48(1):1-7.
[21] 徐淑萍，郭宇，王双，等.NLoS环境下基于UWB的定位算法研究[J].计算机仿真，2023,40(4):439-443.
基本信息:
中图分类号:TN925;TP18
引用信息:
[1]柯希,孙洁.基于EGWO-LSTM-LSM算法的TOF定位研究[J].无线电工程,2024,54(07):1767-1778.
基金信息:
河北自然科学基金和重点基础研究专项(E2019209492)~~
2023-10-17
2023-10-17
2023-10-17