Dataset underlying the research on physical fatigue detection in running using inertial measurement units (IMUs)

This dataset contains data collected during fatigue detection experiments in running using IMUs.Subjects underwent a fatiguing protocol consisting of three distinct consecutive runs on an athletic track:1. The first run consisted of a 4000 m run (10 laps) at a constant speed, determined as 100% of the average speed of the subject during the best performance in the previous year on a 5 to 10 km race;2. The second run was performed according to a fatiguing protocol. The speed in this fatiguing protocol started at the same level of the first run and increased progres-sively of by 0.2km/h every 100 m. Perceived fatigue was assessed by means of a Borg Rating of Perceived Extertion (RPE) Scale (min-max score 6-20) [20], asked to the runner every 100 m. The fatiguing protocol was terminated once the RPE was higher than 16 (RPE between hard and very hard) , or, if such requirement was not met, after 1200m;3. The third run consisted of a 1200m run (3 laps), in which speed was kept constant and equal to the first 4000 m run.<br><br>pXXX_XXX_0-2K: contains the Segment and Joint data exported from MVN for the first half of the first runpXXX_XXX_2-4K: contains the Segment and Joint data exported from MVN for the second half of the first runpXXX_XXX_postfatigue1200m: : contains the Segment and Joint data exported from MVN for the third run<br>pXXX_strides: contains the segmented strides from each subject<br><br>TableFeats: contains values used for the machine learning pipeline, after normalization over each single subject

本数据集收录了跑步疲劳检测实验中采集的惯性测量单元（IMU, Inertial Measurement Unit）数据。受试者需在田径跑道上完成连续三次独立的疲劳诱导跑步测试：
1. 首次跑步：以恒定配速完成4000米（共10圈），该配速设定为受试者过往一年5至10公里赛事最佳成绩的平均配速的100%；
2. 第二次跑步采用疲劳诱导方案：初始配速与首次跑步保持一致，此后每完成100米配速逐步递增0.2km/h。每跑完100米即要求受试者填写伯格感知用力量表（Borg Rating of Perceived Exertion, RPE，评分区间6-20）[20]，以评估其主观疲劳程度。当RPE评分高于16（对应“费力”至“极费力”区间）时终止该疲劳诱导方案；若未达到该阈值，则在完成1200米后强制结束测试；
3. 第三次跑步：以与首次跑步完全一致的恒定配速完成1200米（共3圈）。

`pXXX_XXX_0-2K`：存储首次跑步前半段从MVN导出的节段与关节运动数据；
`pXXX_XXX_2-4K`：存储首次跑步后半段从MVN导出的节段与关节运动数据；
`pXXX_XXX_postfatigue1200m`：存储第三次跑步从MVN导出的节段与关节运动数据；
`pXXX_strides`：存储每位受试者的分段步幅数据；
`TableFeats`：存储经单受试者自身归一化处理后，用于机器学习流水线的特征数值。