基于深度学习的地磁与PDR融合的室内定位算法研究

Research on deep learning-based geomagnetic and PDR fusion algorithm for indoor positioning

导出

摘要为解决地磁定位精度低和PDR存在累计误差的问题,该文提出一种深度学习模型分层优化的多源融合定位方法。首先,构建双层CNN-GRU-Attention模型,利用第一层模型提取地磁特征并进行指纹匹配,随后,构建基于模糊控制的阈值确定模型,以获得精确步数估计结果;最后,利用第二层模型做决策层融合回归预测以获地磁/PDR融合定位结果。实验结果表明,该阈值确定模型可使峰值检测法准确度达97%以上;所构建地磁定位模型性能明显优于传统模型,地磁/PDR融合定位方法平均定位误差为0.662 m,较之单一地磁与PDR定位结果精度分别提升51.04%和74.78%,定位效果显著优于单一定位方式,可获得更为准确的定位轨迹。 To address the issues of low positioning accuracy of geomagnetism and the cumulative error existing in PDR,a multi-source fusion positioning method based on hierarchical optimization of a deep learning model is proposed in this paper.Firstly,a two-layer CNN-GRU-Attention model is constructed.The first layer model is utilized to extract geomagnetic features and conduct fingerprint matching.Subsequently,a threshold determination model based on fuzzy control is established to obtain precise step estimation results.Finally,the second layer model is employed as the decision layer for fusion and regression prediction to obtain the geomagnetism/PDR fusion positioning result.The experimental results indicate that this threshold determination model enables the accuracy of the peak detection method to exceed 97%;the performance of the constructed geomagnetic positioning model is significantly superior to traditional models.The average positioning error of the geomagnetism/PDR fusion positioning method is 0.662 m,and the accuracy is improved by 51.04%and 74.78%respectively compared to the positioning results of single geomagnetism and PDR.The positioning effect is significantly better than that of single positioning methods,and a more accurate positioning trajectory can be obtained。

作者武文静余学祥韩雨辰朱平龙文江 WU Wenjing;YU Xueriang;HAN Yuchen;ZHU Ping;LONG Wenjiang(School of Geomatics,Anhui University of Science and Technology,Huainan,Anhui 232001,China;Urban 3D Real Scene and Intelligent Security Monitoring Joint Laboratory of Anhui Province,Huainan,Anhui 23200l,China;Key Laboratory of Aviation-Aerospace-Ground Cooperative Monitoring and Early Warning of Coal Mining-Induced Disasters of Anhui Higher Education Institutes,Anhui University of Science and Technology,Huainan,Anhui 232001,China;Coal Industry Engineering Research Center of Mining Area Environmental and Disaster Cooperative Monitoring,Anhui University of Science and Technology,Huainan,Anhui 232001,China;School of Earth and Environment,Anhui University of Science and Technology,Huainan,Anhui 2320ol,China;School of Surveying and Mapping,Beijing University of Civil Engineering and Architecture,Beijing 102616,China)