A UWB Radar and Machine Learning-Based Tool for Detecting Victims Through Foliage in Search and Rescue Operations

Project Description During our research in University of West Attica (UniWA) we addressed the problem of victim detection through foliage in Search and Rescue operations. For this purpose, a dataset of respiration signal sessions in the field was collected using a proposed tool consiting of a UWB pulsed radar system, and then these data fed a machine learning tool to enhance FR's operations by providing predictions about human presence behind foliage. In addition, two anemometer sensors were used to record wind data, and a respiration belt was employed to obtain the ground truth measurements about the subject's respiration rate. The setup for each session was the same. The UWB radar [1] was mounted on tripod facing the foliage, the subject was located behind the foliage wearing a respiration belt [4] for breath recording. On the same tripod two anemometers [2],[3] were placed so a comprehesive image of the wind condiditon during the session could be obtained. These sensors were connected to a laptop via USB, about 3 meters away. The distance between the tripod and the foliage was fixed at 1 meter. Foliage (mostly bushes and small olive trees) had length varying from 1 to 3 meters and the subject (in case of presence session) was from 0.5 to 3 meters away from the foliage. In total we never exceeded the 9.2 meters range (unambiguous range) limit of the radar. Dataset Description The dataset consists of 268 sessions of radar, wind and respiration belt data, of which 141 sessions correspond to human presence and 127 to human absence. Each session has a duration of 150 seconds, thus amounting to approximately 6 hours of data for human presence and approximately 5.5 hours of data for human absence. Dataset Contents Each session folder is given an individual name X = posixtime; this name designates the exact time (in posixtime format) when the session was started. For example, in the dataset preview below there can be seen one folder named "1688457913"; this folder corresponds to the measurement session that was initiated exactly on 1688457913 in posixtime format (in this example, X = 1688457913). Furthermore, for the "X" posixtime-named folder, there are the following subfolders and files: 1. One subfolder named Workspaces_X, containing: Files named "Workspace_k.mat", where k the number of the created workspaces containing radar signal recording at 16 FPS. A file named "settings.mat", containing the device settings and the session's distances regarding topology. A file named "windData_original.mat", containing the original data from anemometer sensors saved from the data stream at 4 FPS, provided from a microcontroller followed RS485 protocol. 2. Two files containing the raw data recorded from the respiration belt (only for folders corresponding to human presence and for which a respiration belt was used for obtaining the ground truth measurements of the subject's respiration data.) The "1_YY_MM_DD_HH_MM_general.csv", contains the timestamp in datetime of the sensor and the Android device, the heart rate estimation, the mean breaths per minute and the included IMU belt sensor measurement. The "1_YY_MM_DD_HH_MM_wave.csv", contains the timestamp in datetime of the sensor and the Android device, and 18 values (FPS) of the strain gauge sensor changes from the respiration belt. 3. A file named "X.xlsx", containing the concatenation of the workspaces of the radar signal. 4. A file named "windData_X.csv", containing the synchronized data of anemometer sensors with radar data. 5. A file named "BeltWfm_X.xlsx", containing the synchronized data of respiration belt with radar data (only for folders corresponding to human presence and for which a respiration belt was used for obtaining the ground truth measurements of the subject's respiration data). Proposed Tool COTS components SLMX4 UWB pulse radar [1] Wind Speed [2] and Direction [3] sensors Wind data recording equipment (UART TTL to RS485 Converter, MT3608 DC/DC converter, Arduino) Respiration belt [4]项目描述 本研究由西阿提卡大学（University of West Attica, UniWA）开展，针对搜救行动中透过植被探测受困者的难题展开攻关。为实现该目标，我们采用自研工具（由超宽带脉冲雷达（UWB pulsed radar）系统构成）采集野外呼吸信号会话数据集，随后将该数据集输入机器学习工具，通过提供植被后方人员存在的预测结果，提升FR作业效能。此外，本研究使用两个风速传感器记录风速数据，并采用呼吸带获取受试者呼吸频率的真值测量结果。 所有会话的实验设置均保持一致：超宽带脉冲雷达[1]安装于三脚架上，朝向植被区域；受试者位于植被后方，佩戴呼吸带[4]以记录呼吸信号。同一三脚架上还安装了两个风速传感器[2][3]，以全面采集会话期间的风速状况。上述传感器通过USB接口连接至约3米外的笔记本电脑。三脚架与植被之间的距离固定为1米。植被（以灌木丛和小型橄榄树为主）的覆盖长度范围为1至3米；存在受试者的会话中，受试者与植被的距离为0.5至3米。整体测试未超出该雷达9.2米的无模糊测距范围限制。 数据集说明 本数据集共包含268组雷达、风速及呼吸带数据会话，其中141组对应有受试者场景，127组对应无受试者场景。每组会话时长为150秒，据此统计，有受试者场景的数据总时长约6小时，无受试者场景的数据总时长约5.5小时。 数据集内容 每组会话的文件夹以X=POSIX时间（POSIX time）格式命名，该名称表示会话启动时的精确POSIX时间。例如，本数据集预览中可见一个名为"1688457913"的文件夹，其对应于以POSIX时间格式1688457913启动的测量会话（本示例中X=1688457913）。此外，以POSIX时间命名的文件夹包含以下子文件夹与文件： 1. 名为Workspaces_X的子文件夹，其中包含： - 名为"Workspace_k.mat"的文件，其中k为工作区编号，文件内存储了以16帧每秒（FPS）采集的雷达信号记录数据； - "settings.mat"文件，其中存储了设备设置及会话拓扑相关的距离参数； - "windData_original.mat"文件，其中存储了来自风速传感器的原始数据：数据以4帧每秒（FPS）的速率从数据流中保存，由遵循RS485协议的微控制器提供。 2. 两个存储呼吸带原始记录数据的文件（仅适用于有受试者且使用呼吸带获取受试者呼吸数据真值的会话文件夹）： - "1_YY_MM_DD_HH_MM_general.csv"文件：包含传感器与安卓设备的日期时间戳、心率估算值、平均每分钟呼吸次数，以及内置的惯性测量单元（Inertial Measurement Unit, IMU）带传感器测量数据； - "1_YY_MM_DD_HH_MM_wave.csv"文件：包含传感器与安卓设备的日期时间戳，以及呼吸带应变式传感器的18组每秒采样率的变化数据。 3. "X.xlsx"文件：包含所有雷达信号工作区的拼接数据。 4. "windData_X.csv"文件：包含风速传感器与雷达数据的同步后数据。 5. "BeltWfm_X.xlsx"文件：包含呼吸带与雷达数据的同步后数据（仅适用于有受试者且使用呼吸带获取受试者呼吸数据真值的会话文件夹）。 自研商用现货（Commercial Off-The-Shelf, COTS）组件 - SLMX4超宽带脉冲雷达[1] - 风速[2]与风向[3]传感器 - 风速数据采集设备（UART TTL转RS485转换器、MT3608型DC/DC转换器、Arduino开发板） - 呼吸带[4]