GraspIT构建真实与仿真间闭环，生成高精度抓取姿态数据集。
GraspIT: A Dataset Bridging the Sim-to-Real gap and back for Validated Grasping SE(3) Pose Generation
- 通过物理滑移测试在仿真中标注抓取质量，生成连续评分
- 230万候选抓取中83%通过测试，17%为可分级的难负例
- 支持真实与仿真数据双向映射，适合抓取策略学习与验证
稳健抓取新物体需要同时具备逼真RGB-D观测、经物理验证的抓取质量标注，以及仿真与真实世界的可靠桥梁，现有数据集无法兼顾。GraspIT填补此空白：在NVIDIA Isaac Sim中对桌面场景进行四阶段物理滑移测试，基于平行夹爪机械臂生成轨迹可达性检查与连续质量评分，超越力闭合标准。约230万候选抓取中，83%被标记为良好（得分≥0.50）；17%虽通过力闭合但滑移测试失败，提供分级难负例。通过真实↔仿真循环，将标签回投影至100个真实场景。数据集包含约31.6万组带实例掩码、6-DoF位姿、物理属性和评分的6-DoF抓取的RGB-D帧集，覆盖1035个仿真与100个真实场景。所有工具开源并容器化。Isaac Sim内的轨迹规划还支持高分辨率演示流，用于桌面操作策略学习与行为克隆。
原文摘要 · Abstract (English)
Robust robotic grasping of novel objects requires datasets that simultaneously provide photorealistic RGB-D observations, physically validated grasp quality annotations, and a principled bridge between simulation and the real world, which existing datasets lack to provide jointly. \textbf{GraspIT} addresses this gap: tabletop scenes in NVIDIA Isaac Sim are annotated via a four-stage physical slip-test on parallel Franka Panda instances, producing trajectory-reachability checks and continuous quality scores beyond force-closure.Of ${\sim}$2.3M candidates, 83% pass as \emph{good} ($s{\geq}0.50$); the 17% that passed force-closure but failed the slip-test provide graded hard negatives. A Real$\leftrightarrow$Sim loop back-projects these labels onto 100 real-world scenes. The release provides ${\sim}$316k annotated RGBD frame sets across 1035 sim and 100 real scenes, with instance masks, 6-DoF poses, physical object properties, and scored 6-DoF grasps. All tools are open-source and Docker-containerized. The trajectory planning within Isaac Sim further allows streaming of high resolution demonstrations for tabletop manipulation policy learning and behavior cloning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。