指向并击打：方向条件约束下的可变形线性物体动态操作
Point It, Strike It: Direction-Conditioned Dynamic Manipulation of Deformable Linear Objects
浏览论文内容
中文总结 AI 辅助
本文针对可变形线性物体动态操作中目标包含到达方向的问题，提出DeformX2.0仿真器和TRACE数据生成方法，并利用RECAP策略在真实机器人上显著提升击打成功率。
中文摘要 AI 辅助
目标条件化的可变形线性物体动态操作主要将目标指定为绳索末端需要到达的位置。然而，许多任务取决于末端如何到达。因此，我们研究了单次摆动绳索击打，其目标指定了末端在三维空间中的位置和到达方向，覆盖工作空间并针对不同绳索。这具有挑战性，因为绳索动力学难以建模，不存在演示，不同的摆动以不同的可靠性到达同一目标，并且仿真到现实的差距超出了绳索本身。为应对这些挑战，我们扩展了最先进的DLO仿真器DeformX，加入了GPU加速、稳定的Cosserat杆求解器和横流空气动力学模型，从而得到DeformX2.0，其速度提升超过20,000倍。接着，我们提出了TRACE（轨迹根自适应交叉熵），通过从存储的摆动中热启动每个新目标（其末端路径最接近该目标）来生成击打数据。其代价函数惩罚绳索弯曲和末端突然运动，以有利于可重复的摆动。基于这些数据训练的条件流匹配策略在仿真中达到了92.1%的准确率。最后，我们提出了RECAP（残差校准策略），它将仿真器的绳索和装置参数拟合到几次校准摆动中，并通过在仿真中训练的校正策略来调整动作。在真实机器人上，针对三种绳索，RECAP将位置目标在5厘米内的成功率从72%提高到87%，对于同时指定到达方向的目标，在10厘米和10度内的成功率从50%提高到79%。
英文摘要
Goal-conditioned dynamic manipulation of deformable linear objects has mainly specified goals as positions for a rope tip to reach. Many tasks, however, depend on how the tip arrives. We therefore study single-swing rope striking with goals that specify the tip's 3D position and arrival direction, across the workspace and on different ropes. This is challenging because rope dynamics are hard to model, no demonstrations exist, distinct swings reach the same goal with different reliability, and the sim-to-real gap extends beyond the rope. To address these challenges, we extend the state-of-the-art DLO simulator DeformX with GPU acceleration, a stable Cosserat rod solver, and a cross-flow aerodynamic model, yielding DeformX2.0, which is more than $20{,}000\times$ faster. We then propose TRACE (Trace-rooted Adaptive Cross-Entropy), which generates striking data by warm-starting each new target from the stored swing whose tip path passes closest to it. Its cost penalizes rope bending and abrupt tip motion to favor repeatable swings. A conditional flow-matching policy trained on this data reaches 92.1% accuracy in simulation. Finally, we propose RECAP (Residual Calibration Policy), which fits the simulator's rope and rig parameters to a few calibration swings and adapts actions with a correction policy trained in simulation. On a real robot, across three ropes, RECAP raises success within 5cm from 72% to 87% for position goals, and within 10cm and 10° from 50% to 79% for goals that also specify the arrival direction.
发表机构
- Carnegie Mellon University（卡内基梅隆大学）
- Shanghai Jiao Tong University（上海交通大学）
- Harvard University（哈佛大学）
机构由 AI 辅助整理，请以论文原文为准。
