让机械手抓取更稳定：一次生成姿势、接触点和受力，确保物理可行。
EquiDexFlow: Contact-Grounded SE(3)-Equivariant Dexterous Grasp Generative Flows
- 用流匹配模型联合预测抓取姿态、接触点与受力，自动满足摩擦约束。
- 在81个物体上训练后，零摩擦违规，抓取力矩残差最低，性能最优。
- 可适配真实机器人，抓取成功率100%，支持复杂非对称物体抓握。
大多数学习型灵巧抓取生成器将接触力放在后续验证步骤，导致运动学可行的姿态仍可能违反稳定抓取条件。本文提出EquiDexFlow，一种SE(3)等变的流匹配模型，从物体点云中联合预测手腕姿态、关节角度、指尖接触点、表面法向及接触力。模型通过构造将接触点投影到物体表面、力约束于库仑摩擦锥，无需损失惩罚即可保证放置与摩擦合规性。理论证明并实证验证了端到端的SE(3)等变性，在200次旋转下腕部残差低于0.04°，关节偏差精确为零。在81个物体共8,100个力闭合抓取数据上训练，针对16自由度Allegro手，模型实现零摩擦违规，综合得分最高，力矩残差最低。通过每指逆运动学重定向至16自由度LEAP手，优化后的关节均至少位于执行器包络内5%以上，同时保持力平衡。在真实机器人上，重定向后的抓取成功完成所有6个测试物体的开环拾取与保持实验，每个非对称物体在标准位姿和120°共旋转位姿下均成功。视频、代码与检查点见https://equidexflow.github.io。
原文摘要 · Abstract (English)
Most learned dexterous grasp generators relegate contact forces to a downstream verification step, so a kinematically-plausible pose can still violate the conditions for a stable physical grasp. We address this with EquiDexFlow, an SE(3)-equivariant flow-matching model that jointly predicts wrist pose, joint angles, fingertip contacts, surface normals, and contact forces from an object point cloud. Our architecture projects contacts onto the object surface and forces into the Coulomb friction cone by construction, so placement and friction compliance hold without loss penalties. We prove end-to-end SE(3) equivariance and verify it empirically over 200 rotations, with wrist residuals below $0.04^\circ$ and exactly zero joint deviation. Trained on 8,100 force-closure grasps across 81 objects for the 16-DoF Allegro Hand, our model achieves zero friction violations, the best composite score, and the lowest wrench residual among all ablation variants. We retarget decoded fingertip contacts to a 16-DoF LEAP Hand via per-finger inverse kinematics, and our hardware-feasible refinement places every joint at least 5% inside its actuator envelope while preserving wrench balance. On the physical robot, retargeted EquiDexFlow-decoded grasps complete open-loop pick-and-hold trials on all six test objects, with every asymmetric object succeeding at both the canonical pose and a $120^\circ$ co-rotation. Videos, code, and checkpoints are available at https://equidexflow.github.io.
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。