A multi-sensory dataset for the activities of daily living

The dataset contains multiple instances of 9 Activities of Daily Living (ADL)-related actions namely Walk, Sit Down, Stand Up, Open Door, Close Door, Pour Water, Drink Glass, Brush Teeth and Clean Table. Each of the 10 volunteers performed each activity at least 14 times, with the notable exception of the walking activity that has been performed 40 times, in different sequences and alternating the used hand. For each volunteer, the dataset contains 7 CSV files, i.e., one file for each of the 6 IMU sensors worn by the volunteer on different body parts, as described in Figure 1, namely left lower arm (lla.csv), left upper arm, (lua.csv), right lower arm (rla.csv), right upper arm (rua.csv) and right thigh (rt.csv). Each file contains the overall sequence recorded during the experiment. The first column contains a label "qags" indicating the type of recorded data (quaternions, acceleration, angular velocity). The next column is the time-stamp in milliseconds elapsed from 00:00:0.000 AM (with a 30 milliseconds sampling time). The next four columns are the quaternions (with a resolution of 0.0001 ). Following them, we have three columns with the accelerations along the x, y and z axes (with a 0.1 mG resolution). The last three columns refer to the angular velocity about the x, y and z axes (with a 0.01 dps resolution). The last CSV file (annotation.csv) contains the data labelling. The first two columns of this file contain the time in the format hh.mm.ss.000 (current day time) and in milliseconds elapsed from 00:00:0.000 AM. All the remaining columns are organised as couples where the first element represents the scope of the labelling and the second indicates whether the labelled activity is starting or ending. In the annotation file, there are four different labelling scopes. • “BothArms”: all instances of each activity are labelled independently of which arm has been used; • “RightArm”: are labelled the activity instances using only the right arm, or in case they belong to the Walk, Sit Down or the Stand Up activities; • “LeftArm”: are labelled the activity instances using only the left arm, or in case they belong to the Walk, Sit Down or the Stand Up activities; • “Locomotion”: in this scope are labelled only the instances of Walk, Sit Down and Stand Up. Finally, the last two columns report a session ID. There are four different sessions characterised by the order in which the activities are performed, and by the used arm (see also Table 3), and whether the session starts or ends. The videos recorded during the experiments have been only used for labelling purposes, and they are not published. Together with the dataset, we provide a MATLAB script named TimeStampExtraction.m that extract from the annotation and the data files, for each volunteer and for each sensor, the time stamp associated with the start and end of each ADL.

本数据集包含9项日常生活活动（Activities of Daily Living, ADL）相关动作的多组样本，具体为行走、坐下、站起、开门、关门、倒水、持杯饮水、刷牙及擦拭桌面。10名志愿者均完成了每项活动至少14次，其中行走动作的样本量达40次，所有动作均以不同序列完成，并交替使用左右手。每位志愿者对应7个CSV格式文件，其中6个文件对应其佩戴于不同身体部位的6个惯性测量单元（Inertial Measurement Unit, IMU）传感器，具体部位如图1所示，分别为左前臂（lla.csv）、左上臂（lua.csv）、右前臂（rla.csv）、右上臂（rua.csv）及右大腿（rt.csv），剩余1个为标注文件annotation.csv。每个传感器文件包含实验全程记录的序列数据：第一列为标签“qags”，用于标识记录数据的类型（四元数、加速度、角速度）；第二列为自00:00:0.000起的毫秒级时间戳，采样间隔为30毫秒；后续四列为四元数数据，分辨率为0.0001；再之后的三列为沿x、y、z轴的加速度数据，分辨率为0.1毫高斯（mG）；最后三列为绕x、y、z轴的角速度数据，分辨率为0.01度每秒（dps）。
annotation.csv包含标注数据，其前两列分别为以hh.mm.ss.000格式表示的当日时间，以及自00:00:0.000起的毫秒级时间戳；剩余列均以列对形式组织，第一元素表示标注范围，第二元素表示对应动作的起始或结束状态。本标注文件共包含4种标注范围：
1. "BothArms"：不限制使用手臂的所有活动实例均会被标注；
2. "RightArm"：仅使用右手完成的活动实例会被标注，但若活动属于行走、坐下或站起，则无论手臂使用情况均会被标注；
3. "LeftArm"：仅使用左手完成的活动实例会被标注，但若活动属于行走、坐下或站起，则无论手臂使用情况均会被标注；
4. "Locomotion"：仅对行走、坐下及站起的活动实例进行标注。
最后两列为会话ID，共存在4种会话类型，其区分依据为活动执行顺序、使用的手臂（详见表3）以及会话的起始/结束状态。实验过程中录制的视频仅用于标注用途，未随数据集一同发布。本数据集附带一款名为TimeStampExtraction.m的MATLAB脚本，可针对每位志愿者及每个传感器，从标注文件与数据文件中提取与每项日常生活活动起止时刻对应的时间戳。