PDR algorithm for precise positioning of underground personnel based on LSTM personalized step size estimation
-
摘要: 针对传统的行人航位推算(PDR)算法由于步长和航向累积误差导致定位精度较低，不能满足井下人员精准定位需求的问题，提出了一种基于长短时间记忆网络(LSTM)个性化步长估计的井下人员精准定位PDR算法。首先采集井下人员运动中的加速度、陀螺仪惯性信息，解算每一步运动距离构建步长数据，通过离线训练获得井下人员个性化步长估计LSTM模型；然后在在线预测阶段通过矿用本安智能手机实时采集加速度、陀螺仪、地磁等井下人员运动数据，分别采用步伐检测算法、个性化步长估计模型获得井下人员运动步伐及每一步的步长，利用卡尔曼滤波融合航向估计算法获得航向角；最后根据步长估计和航向角预测井下人员当前位置。在内蒙古鄂尔多斯市高头窑煤矿采集井下人员运动数据进行试验，结果表明：基于LSTM个性化步长估计的井下人员精准定位PDR算法对井下人员运动中的步伐检测精度为96.5%，步长预测精度为90%；在井下真实环境中的相对定位误差为2.33%，提高了煤矿井下人员定位的精度。Abstract: The traditional pedestrian dead reckoning (PDR) algorithm has low positioning precision due to the accumulated errors of step size and heading, which can not meet the requirements of precise positioning of underground personnel. In order to solve the problem, a PDR algorithm for precise positioning of underground personnel based on long short-term memory (LSTM) personalized step size estimation is proposed. Firstly, the acceleration and gyroscope inertia information in the movement of underground personnel is collected, and the movement distance of each step is calculated to construct step size data. The LSTM model of personalized step size estimation of the underground personnel is obtained through off-line training. Secondly, in the online prediction stage, the underground personnel movement data such as acceleration, gyroscope and geomagnetism are collected in real-time through the mine intrinsically safe smart phone. The underground personnel movement step and step size of each step are obtained by using the step detection algorithm and personalized step size estimation model respectively. The heading angle is obtained by using the Kalman filtering and heading estimation algorithm. Finally, the current position of underground personnel is predicted according to step size estimation and heading angle. In Inner Mongolia Ordos Gaotouyao Coal Mine, the underground personnel movement data is collected for testing, and the results show as follows. The PDR algorithm for precise positioning of underground personnel based on LSTM personalized step size estimation has a step detection precision of 96.5% and a step size prediction precision of 90%. The algorithm has a relative positioning error of 2.33% in the real underground environment, which improves the personnel positioning precision in coal mine.
-
图 2 井下人员个性化步长估计LSTM模型框架
Figure 2. LSTM model framework of underground personnel personalized step size estimation
图 5 井下人员个性化步长估计LSTM模型的输入
Figure 5. Input of LSTM model of underground personnel personalized step size estimation
图 6 井下人员个性化步长估计LSTM模型的结构参数
Figure 6. Structural parameters of LSTM model of underground personnel personalized step size estimation
图 8 不同步长估计算法误差分布对比
Figure 8. Comparison of error distribution of different step size estimation algorithms
图 11 传统PDR算法与基于LSTM个性化步长估计的PDR算法在煤矿井下行走路径对比
Figure 11. Comparison of walk routes of traditional PDR algorithm and PDR algorithm based on LSTM personalized step size estimation in coal mine
表 1 井下人员个性化步长估计LSTM模型的超参数
Table 1. Super parameter of LSTM model of underground personnel personalized step size estimation
批量大小 激活函数 优化器 学习率 迭代次数 早停次数 损失函数 128 ReLU Adam 0.001 500 50 MSE 
下载: 导出CSV
表 2 步伐检测算法结果
Table 2. Results of step detection algorithm
试验
次数试验者1 试验者2 试验者3 误检步数 准确率/% 误检步数 准确率/% 误检步数 准确率/% 1 3 94 1 98 2 96 2 2 96 0 100 6 88 3 1 98 0 100 1 98 4 0 100 0 100 7 86 5 1 98 1 98 1 98
下载: 导出CSV
-
[1] 李晨辉, 甄杰, 祝会忠, 等. 复杂环境下的超宽带高精度定位算法[J]. 测绘科学,2020,45(1):4-10.LI Chenhui, ZHEN Jie, ZHU Huizhong, et al. Ultra wideband high precision positioning algorithm in complex environment[J]. Science of Surveying and Mapping,2020,45(1):4-10. [2] YU N, ZHAN X, ZHAO S, et al, A precise dead reckoning algorithm based on bluetooth and multiple sensors[J]. IEEE Internet Things Journal, 2018, 5(1): 336-351. [3] 孙哲星. 煤矿井下人员精确定位方法[J]. 煤炭科学技术,2018,46(3):130-134.SUN Zhexing. Accurate positioning method of underground personnel in coal mine[J]. Coal Science and Technology,2018,46(3):130-134. [4] 孙延鑫, 毛善君, 苏颖, 等. 改进的井下人员定位PDR算法研究[J]. 工矿自动化,2021,47(1):43-48.SUN Yanxin, MAO Shanjun, SU Ying, et al. Research on improved PDR algorithm for underground personnel positioning[J]. Industry and Mine Automation,2021,47(1):43-48. [5] 郭娅婷, 杨君, 甘露. 基于改进PDR与RSSI融合的定位算法[J]. 传感技术学报,2020,33(7):1027-1032. doi: 10.3969/j.issn.1004-1699.2020.07.016GUO Yating, YANG Jun, GAN Lu. Localization algorithm based on improved PDR and RSSI fusion[J]. Journal of Sensor Technology,2020,33(7):1027-1032. doi: 10.3969/j.issn.1004-1699.2020.07.016 [6] WEINBERG H. Using the ADXL202 in pedometer and personal navigation applications[EB/OL]. (2016-09-23) [2021-06-15].https://www.docin.com/p-1743985325.html. [7] KIM J W, HAN J J, HWANG D H, et al. A step, stride and heading determination for the pedestrian navigation system[J]. Journal of Global Positioning Systems,2004,3(1/2):273-279. [8] SCARLETT J. Enhancing the performance of pedometers using a single accelerometer [EB/OL]. (2020-06-02) [2021-06-15].https://www.86ic.net/qiche/xinnengyuan/44427.html. [9] HAN G, GROVES P D . Context determination for adaptive navigation using multiple sensors on a smartphone [C]//Ion Gnss, 2016: 12-16. [10] HANNINK J, KAUTZ T, PASLUOSTA C, et al. Mobile stride length estimation with deep convolutional neural networks[J]. IEEE Biomedical and Health Informatics,2018,22(2):354-362. doi: 10.1109/JBHI.2017.2679486 [11] 张荣辉, 贾宏光, 陈涛, 等. 基于四元数法的捷联式惯性导航系统的姿态解算[J]. 光学精密工程,2008,16(10):1963-1970. doi: 10.3321/j.issn:1004-924X.2008.10.029ZHANG Ronghui, JIA Hongguang, CHEN Tao, et al. Attitude calculation of strapdown inertial navigation system based on quaternion method[J]. Optics and Precision Engineering,2008,16(10):1963-1970. doi: 10.3321/j.issn:1004-924X.2008.10.029 [12] 肖宇. 基于互补滤波算法的四旋翼飞行器姿态和高度解算[J]. 工业控制计算机,2016,29(10):94-96. doi: 10.3969/j.issn.1001-182X.2016.10.045XIAO Yu. Attitude and height estimation of quad-rotor aircraft based on complementary filter[J]. Industrial Control Computer,2016,29(10):94-96. doi: 10.3969/j.issn.1001-182X.2016.10.045 [13] YADAV N, BLEAKLEY C. Accurate orientation estimation using AHRS under conditions of magnetic distortion[J]. Sensors,2014,14(11):20008-20024. doi: 10.3390/s141120008 [14] WANG Qu, LUO Haiyong, YE Langlang, et al. Personalized stride-length estimation based on active online learning[J]. IEEE Internet of Things Journal,2020,7(6):4885-4897. doi: 10.1109/JIOT.2020.2971318 [15] 毕京学, 汪云甲, 曹鸿基, 等. 一种波峰波谷检测的智能手机计步算法[J]. 中国惯性技术学报,2020,28(3):287-292.BI Jingxue, WANG Yunjia, CAO Hongji, et al. A step counting algorithm for smartphone with peak-valley detection[J]. Journal of Chinese Inertial Technology,2020,28(3):287-292. -
点击查看大图
计量
- 文章访问数: 266
- HTML全文浏览量: 157
- PDF下载量: 44
- 被引次数: 0
