灵巧作业臂-手机器人的动作基元库数据集

本数据集于2022年11月至2025年10月在重庆大学完成，基于 UR5 机械臂 + 灵巧手在真实多场景、多任务条件下实地采集。数据集涵盖灵巧臂-手系统对抓、捏、握、拧、旋、推、拉、按、插、拔等10种基础操作模式的数据。实验针对杯子、衣物、线缆等典型物体开展可重复测试，同步采集关节角度、灵巧手力觉数据及多模态触觉数据。本数据集可用于动作基元理解、抓取分析、力觉融合、多模态学习以及机器人操作任务的研究和模型训练。This dataset was collected at Chongqing University from November 2022 to October 2025, using a UR5 robotic arm paired with a dexterous hand under realistic multi-scenario and multi-task conditions. The dataset includes data covering 10 basic operation modes of the dexterous arm-hand system, namely grasping, pinching, holding, twisting, rotating, pushing, pulling, pressing, inserting and pulling out. Experiments were conducted on typical objects such as cups, garments and cables for repeatable testing, with joint angles, dexterous hand force data and multimodal tactile data synchronously collected. This dataset can be used for research and model training in action primitive understanding, grasping analysis, force fusion, multimodal learning and robotic manipulation tasks.

搜集汇总

背景与挑战

背景概述

该数据集于2022年11月至2025年10月在重庆大学采集，基于UR5机械臂和灵巧手在真实多场景下，针对杯子、衣物等典型物体，记录了抓、捏、握等10种基础操作模式的关节角度、力觉和多模态触觉数据。它适用于动作基元理解、抓取分析、力觉融合及机器人操作任务的研究与模型训练。