A multimodal EMG and IMU dataset for assessing the quality of exercises designed for spatially constrained environments

We present a multimodal exercise dataset collected to support autonomous feedback systems for training in spatially constrained environments. Twenty healthy adults performed a structured set of whole-body exercises in limited space. The dataset includes surface EMG from four muscles, IMU kinematics (quaternions), wrist heart rate, and expert annotations of movement quality against predefined biomechanical criteria. Raw signals were filtered, segmented, synchronized, and EMG envelopes were derived. This resource enables development and validation of machine-learning models for automated assessment of exercise quality when professional supervision is not available.本研究构建了一个多模态运动数据集，旨在为空间受限环境下的训练自主反馈系统提供支撑。20名健康成年受试者在有限空间内完成了一套结构化全身运动训练序列。该数据集包含四类数据：四块肌肉的表面肌电（surface EMG）信号、惯性测量单元（IMU）采集的运动学数据（以四元数形式存储）、腕部心率数据，以及基于预定义生物力学标准的运动质量专家标注结果。原始信号已完成滤波、分段与同步处理，并提取了肌电包络特征。该数据集可用于开发并验证无需专业人员在场即可自动评估运动质量的机器学习模型。