面向应急救援的BP神经网络人员定位-追踪-静止报警系统

doi:10.16265/j.cnki.issn1003-3033.2026.04.0177

摘要/Abstract

摘要：

为及时救治火场中被困人员,提出一种耦合反向传播(BP)神经网络和双侧双向测距(DS-TWR)技术的室内人员定位-追踪-及静止报警系统。该系统利用实验室虚拟仪器工程工作台(LabVIEW)整合超宽带(UWB)的DS-TWR技术和BP神经网络,通过BP神经网络学习多径效应与非视距传播的误差特征,修正DS-TWR误差,精确定位人员位置,同步记录行动轨迹;通过计算人员在指定时间范围内的移动距离来评估运动状态,设定位移与时间阈值构建报警机制,当检测到人员静止超过安全阈值时系统自动触发报警。结果表明:该系统在正常及金属障碍电磁干扰环境下,静态及动态定位误差可达厘米级,具有定位精度高、稳定性好的优势,且轨迹绘制精准,人员静止报警信息准确,响应延迟控制在较低水平。

关键词: 应急救援, 反向传播(BP)神经网络, 人员定位-追踪-静止报警, 双侧双向测距(DS-TWR), 实验室虚拟仪器工程工作台(LabVIEW)

Abstract:

To ensure timely rescue of trapped personnel in fire scenarios, a personnel positioning-tracking-stationary alarm system was proposed by combining BP neural networks and DS-TWR technology. The system was developed based on the LabVIEW platform, where the BP neural network learned the error patterns of multipath effects and non-line-of-sight propagation, corrected DS-TWR ranging errors, and achieved precise personnel positioning and trajectory tracking. By calculating the movement distance within a defined time, the system evaluated personnel motion status and established thresholds of displacement and time to trigger alarms, which automatically activated alerts when a person remained stationary beyond the safety threshold. Test results show that the system achieves centimeter-level static and dynamic positioning accuracy under both normal and metal/electromagnetic interference environments. It features high positioning precision and favorable stability, enables accurate trajectory generation, delivers reliable static personnel alarms, and maintains a low response latency.

Key words: emergency rescue, back propagation (BP) neural network, personnel positioning-tracking-stationary alarm, double-sided two-way ranging (DS-TWR), laboratory virtual instrument engineering workbench (LabVIEW)

中图分类号:

X924.4安全控制技术

王丽, 王喆, 关文玲, 张嘉琪, 刘超林, 孟玉莹. 面向应急救援的BP神经网络人员定位-追踪-静止报警系统[J]. 中国安全科学学报, 2026, 36(4): 204-210.

Wang Li, Wang Zhe, Guan Wenling, Zhang Jiaqi, Liu Chaolin, Meng Yuying. BP neural network-based personnel positioning-tracking-stationary alarm system for emergency rescue[J]. China Safety Science Journal, 2026, 36(4): 204-210.

图/表 9

图1

图2

图3

表1

图4

图5

图6

图7

表2

参考文献 26

[1]	国务院事故调查组. 北京丰台长峰医院“4·18”重大火灾事故调查报告[R], 2023.
[2]	应急管理部. 江西新余佳乐苑小区“1·24”特别重大火灾事故调查报告公布[EB/OL]. (2024-09-21). https://www.mem.gov.cn/xw/yjglbgzdt/202409/t20240921_499688.shtml.
[3]	新华社. 河北承德国恩老年公寓“4·8”重大火灾事故调查报告发布[EB/OL]. (2025-12-30). https://news.sina.com.cn/c/2025-12-30/doc-inhequpm5696365.shtml.
[4]	四川省人民政府. 四川自贡九鼎大楼“7·17”重大火灾事故调查报告[R], 2025.
[5]	王纯. 面向应急场景的人员检测与定位技术研究[D]. 长沙: 湖南大学, 2023.
	Wang Chun. Research on personnel detection and localization technology for emergency rescue scenario[D]. Changsha: Hunan University, 2023.
[6]	莫轶安. 消防救援现场人员定位技术对比优化分析[J]. 武警学院学报, 2021, 37(6):36-39.
	Mo Yian. Comparative and optimal analysis of positioning technologies for fire response operations[J]. Journal of the Armed Police Academy, 2021, 37(6):36-39.
[7]	Faheem Z, Athanasios G, Kin K L. A survey of indoor localization systems and technologies[J]. IEEE Communications Surveys and Tutorials, 2019, 21(3):2568-2599. doi: 10.1109/COMST.2019.2911558
[8]	张淼, 王孝军, 雷经发, 等. 轻量化神经网络结合深度相机的矿工目标检测与定位[J]. 中国安全科学学报, 2025, 35(3):115-124. doi: 10.16265/j.cnki.issn1003-3033.2025.03.0863
	Zhang Miao, Wang Xiaojun, Lei Jingfa, et al. Lightweight neural network combined with depth camera for miner target detection and localization[J]. China Safety Science Journal, 2025, 35(3): 115-124. doi: 10.16265/j.cnki.issn1003-3033.2025.03.0863
[9]	Wang Huan, Wang Guijin, Li Xianting. Image-based occupancy positioning system using pose-estimation model for demand-oriented ventilation[J]. Journal of Building Engineering, 2021, 39: DOI: 10.1016/j.jobe.2021.102220.
[10]	竹记. 应急救援自主定位系统的设计与实现[D]. 北京: 北京邮电大学, 2020.
	Zhu Ji. Design and implementation of emergency rescue autonomous positioning system[D]. Beijing: Beijing University of Posts and Telecommunications, 2020.
[11]	徐盛良, 鲁照权, 周永燕, 等. 基于信号强度与超声波测距的室内定位系统[J]. 仪表技术与传感器, 2019(1): 91-96.
	Xu Shengliang, Lu Zhaoquan, Zhou Yongyan, et al. Indoor positioning system based on signal strength and ultrasonic distance measurement[J]. Instrument Technique and Sensor, 2019(1): 91-96.
[12]	陈沛宇, 袁勤政, 戴鹏飞, 等. 多技术融合的室内无线定位方法发展综述[J]. 导航定位学报, 2022, 10(3):9-13.
	Chen Peiyu, Yuan Qinzheng, Dai Pengfei, et al. Overview of the development of indoor wireless positioning methods based on multi-technology integration[J]. Journal of Navigation and Positioning, 2022, 10(3): 9-13.
[13]	Zhou Hang, Cong Haiyong, Wang Yuanyuan, et al. A computer-vision-based deep learning model of smoke diffusion[J]. Process Safety and Environmental Protection, 2024, 187: 721-735. doi: 10.1016/j.psep.2024.04.123
[14]	Gomes E L, Fonseca M, Lazzaretti A E, et al. Clustering and hierarchical classification for high-precision RFID indoor location systems[J]. IEEE Sensors Journal, 2022, 22(6): 5141-5149. doi: 10.1109/JSEN.2021.3103043
[15]	洪利, 武有文, 李亚南, 等. 面向生命探测的WiFi定位系统研究[J]. 自然灾害学报, 2017, 26(5):23-29.
	Hong Li, Wu Youwen, Li Ya'nan, et al. Research on WiFi positioning system for life detection[J]. Journal of Natural Disasters. 2017, 26(5): 23-29.
[16]	Xu Jingao, Yang Zheng, Chen Hengjie, et al. Embracing spatial awareness for reliable WiFi-based indoor location systems[C]. 2018 IEEE 15^th International Conference on Mobile Ad Hoc and Sensor Systems (MASS), 2018: 281-289.
[17]	钟炜, 于佳怡. 智慧城市背景下的火灾实时疏散系统[J]. 中国安全科学学报, 2023, 33(2): 179-184. doi: 10.16265/j.cnki.issn1003-3033.2023.02.2798
	Zhong Wei, Yu Jiayi. Real-time fire evacuation system under smart city background[J]. China Safety Science Journal, 2023, 33(2): 179-184. doi: 10.16265/j.cnki.issn1003-3033.2023.02.2798
[18]	李锋. 煤矿井下精确定位技术现状及应用[J]. 工矿自动化, 2023, 49(增刊1):44-46.
	Li Feng. Present situation and application of underground precise positioning technology[J]. Journal of Mine Automation, 2023, 49(S1): 44-46.
[19]	许伟, 郭晓娜, 陈康, 等. 移动蜂窝网络定位技术综述[J]. 湘潭大学学报(自然科学版), 2024, 46(1):44-57.
	Xu Wei, Guo Xiaona, Chen Kang, et al. Overview of mobile cellular network positioning technology[J]. Journal of Xiangtan University (Natural Science Edition), 2024, 46(1):44-57.
[20]	宫培松, 沈欣云, 韩博雯, 等. 深基坑工程人员区域入侵行为预警数字孪生系统[J]. 中国安全科学学报, 2025, 35(8): 33-39. doi: 10.16265/j.cnki.issn1003-3033.2025.08.1471
	Gong Peisong, Shen Xinyun, Han Bowen, et al. Digital twin system for early warning of regional intrusion behavior of personnel in deep foundation pit projects[J]. China Safety Science Journal, 2025, 35(8): 33-39. doi: 10.16265/j.cnki.issn1003-3033.2025.08.1471
[21]	Tu Cheng, Zhang Jiabin, Quan Zhi, et al. UWB indoor localization method based on neural network multi-classification for NLOS distance correction[J]. Sensors and Actuators A: Physical, 2024, 379: 1-13.
[22]	Khaldon A K, Mardeni R, Mohamad Y A, et al. Survey of indoor localization based on deep learning[J]. Computers, Materials & Continua, 2024, 79(2): 3261-3298.
[23]	Yang Hongchao, Wang Yunjia, Chee K S, et al. A novel credibility evaluation and mitigation for ranging measurement in UWB localization[J]. Measurement, 2025, 256: 1-12.
[24]	Wang Penghui, Lian Zengzeng, Núñez-AndrÉs M A, et al. A GCN-GRU-KAN-based framework for UWB 3D localization in adverse geometric configurations[J]. Measurement, 2026, 258: 1-12.
[25]	石娟, 常丁懿, 郑鹏. 基于BP神经网络的建筑工人不安全行为预警模型[J]. 中国安全科学学报, 2022, 32(1): 27-33. doi: 10.16265/j.cnki.issn1003-3033.2022.01.004
	Shi Juan, Chang Dingyi, Zheng Peng. An early warning model of unsafe behaviors of construction workers based on BP neural network[J]. China Safety Science Journal, 2022, 32(1): 27-33. doi: 10.16265/j.cnki.issn1003-3033.2022.01.004
[26]	梁祎. 基于ReLU神经网络的最优函数逼近[D]. 北京: 北京工业大学, 2021.
	Liang yi. Optimal function approximation using ReLU neural networks[D]. Beijing: Beijing University of Technology, 2021.

折数	CEP50/m	CEP95
第1折	0.031	0.068
第2折	0.029	0.064
第3折	0.033	0.071
第4折	0.030	0.066
第5折	0.032	0.069
均值±标准差	0.031 ± 0.001 6	0.068 ± 0.002 7

环境	静态 RMSE/m	动态 RMSE/m	MAE/ m	报警延迟/s
正常	0.05	0.06	0.06	2
金属和电磁干扰	0.07	0.085	0.16	2.6