让机器人在杂乱中自主判断刀片插入的可行空间，提升仓储搬运效率。
Learning 3D Affordances for Blade Insertion in Cluttered Stowing
- 用3D体素场重建刀片可通行区域，直接从场景几何推断空间可行性。
- 在1万次真实仓库操作上训练，覆盖率达0.89，优于传统方法0.71。
- 适合做复杂环境下的机器人空间推理，尤其对无标注数据场景有效。
许多操作任务需要理解自由空间可达性：发现刚性工具可安全通行的三维体积，这与抓取所需的表面接触可达性互补。机器人仓储是典型应用，需将刀片在杂乱布料箱内推开物品以创建插入空间。生产级仓储系统每天生成数百万次此类操作，但传统方法仅依赖单模态数据，将可达性建模为SE(3)位姿分布，本质是错误领域的问题。VulcanVoxel采用空间化推理：基于掩码自编码器在3D占据场中重建刀片占据状态，局部计算每个体素的可行性，并从单模态数据中恢复多模态预测。刀片可达性是空间对象，即由几何可行性定义的3D空间子集。位姿参数缺乏结构，无法判断未观测位置是否可行；标准生成目标（如流匹配）仅能学习执行策略产生的单模态分布，无法恢复几何上的替代方案。模型在10,000个真实仓库堆叠任务中训练，无需人工标注，达到top-5覆盖率为0.89，优于最佳位姿基线的0.71；压缩后的学生模型可在30毫秒内完成从RGB到体素的推理，远快于体素到体素的1.4秒。相关数据集已公开，包含真实刀片插入循环的RGB-D观测与位姿轨迹，地址为https://www.armbench.com/blade_insertion.html。
原文摘要 · Abstract (English)
Many manipulation tasks require reasoning about free-space affordances: discovering volumes where an extended rigid tool can safely navigate, complementary to surface contact affordances for grasping. Robotic stowing is a canonical instance, where a blade must sweep items aside inside cluttered fabric bins to create insertion space. Production stow systems generate millions of such episodes, but standard approaches with unimodal data infer affordances as SE(3) pose distributions, a geometric question asked in the wrong domain. VulcanVoxel keeps inference spatial: a masked autoencoder over 3D occupancy fields reconstructs blade occupancy conditioned on scene geometry, computing feasibility locally at each voxel and recovering multi-modal predictions from unimodal data. Blade affordances are spatial objects, subsets of 3D space defined by geometric feasibility. Pose parameters carry no structure for reasoning whether unobserved placements are feasible, and standard generative objectives including flow matching faithfully learn the unimodal distribution produced by execution policies and cannot recover geometric alternatives. Trained on 10,000 real warehouse stow episodes without human annotation, VulcanVoxel achieves top-5 coverage of 0.89 versus 0.71 for the best pose-based baseline, with a distilled student providing RGB-to-voxel inference in 30 ms. vs. 1.4 s. for voxel-to-voxel. We have released a dataset of real blade insertion cycles with RGB-D observations and pose trajectories at https://www.armbench.com/blade_insertion. html.
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。