MOVMUS-UJI Dataset & ERGOMOVMUS: EMG and kinematics data of the hand in activities of daily living with special interest for ergonomics

A dataset of human hand kinematics and forearm muscle activation collected during the performance of a wide variety of activities of daily living (ADLs) is presented, with tagged characteristics of products and tasks. A total of 26 participants performed 161 ADLs, selected to be representative of common elementary tasks, grasp types, product orientations and performance heights. 105 products were used, being varied regarding shape, dimensions, weight and type (common products and assistive devices). The data were recorded using CyberGlove instrumented gloves on both hands measuring 18 degrees of freedom on each and seven surface EMG sensors per arm recording muscle activity. The products and their arrangement were the same across subjects, and tasks were performed in a guided way. Data of more than 4100 ADLs is presented in this dataset as Matlab structures with full continuous recordings, which may be used in applications such as machine learning or to characterize healthy human hand behaviour. The dataset is accompanied with a custom data visualization application (ERGOMOVMUS) as a tool for ergonomics applications, allowing visualization and calculation of aggregated data from specific task, product and/or subjects’ characteristics. v2.1 includes the following updates: - Statistical summary of the recordings both in .xlsx and .ods file format (v1.1 only included it in .xlsx file format) - Updated experiment details in "MOVMUS-UJI DATASET GUIDE.pdf".

本数据集收录了人类在执行多种日常活动（Activities of Daily Living, ADLs）过程中的手部运动学与前臂肌肉激活数据，并标注了相关产品与任务的特征。共计26名受试者完成了161项日常活动，所选活动均覆盖常见基础任务、握持类型、产品摆放角度与操作高度，具备充分代表性。本次实验共使用105款产品，其形状、尺寸、重量与类型（含普通日用品与辅助器具）均存在差异。数据采集采用双手佩戴CyberGlove测力手套，每只手套可采集18个自由度的运动数据；同时每臂搭载7个表面肌电（surface EMG）传感器，记录肌肉活动信号。所有受试者的实验产品与摆放方式均保持一致，任务均在标准化引导流程下完成。本数据集共包含超过4100项日常活动的连续完整记录，以Matlab结构体格式存储，可用于机器学习或健康人群手部行为特征分析等相关应用。数据集配套自研数据可视化工具ERGOMOVMUS，面向人机工效学应用开发，支持针对特定任务、产品及/或受试者特征的可视化与聚合数据计算。v2.1版本新增以下更新内容：
- 提供.xlsx与.ods两种格式的记录统计摘要文件（v1.1版本仅支持.xlsx格式）
- 更新了"MOVMUS-UJI DATASET GUIDE.pdf"中的实验细节说明。

搜集汇总

背景与挑战

背景概述

MOVMUS-UJI数据集是一个专注于人机工程学的手部运动学和前臂肌电数据集，收集了26名参与者在执行161种日常活动时的数据，涉及105种产品，记录超过4100个活动。数据以连续记录形式提供，适用于机器学习和健康手部行为分析，并附带可视化工具ERGOMOVMUS，便于任务和产品特性分析。