让静态训练的机械臂学会抓动目标，无需重新训练。
Static In, Dynamic Out: Counterfactual Action Augmentation for Moving Object Manipulation
- 用反事实位置替换目标，让动作保持手物相对姿态。
- 在5种运动模式下，成功率比基线提升15%-30%。
- 适合真实场景中目标会移动的抓取任务。
视觉-运动策略在目标静止时表现良好，但实际应用中目标常移动：传送带上零件漂移、水果随风摆动。本文提出静态输入、动态输出（SIDO）方法，使仅用静态示范训练的策略能适应测试时未见的目标运动。核心思路是将动态抓取分解为两步：预测目标未来位置，并到达该位置。SIDO将目标移至反事实未来位置，调整示范动作以保持手物相对姿态，形成目标条件策略。部署时由目标位姿预测器提供未来位置。在三个模拟任务（Mug、Square、Stack）的五种运动模式及两个真实任务（Gantry、Peachtree）中，SIDO均显著优于基线，且保持对静态目标的性能。项目网站：https://sido-staticindynamicout.github.io/。
原文摘要 · Abstract (English)
Visuomotor policies have advanced on manipulation tasks where the target object stays static during execution, but real deployments break this assumption: parts drift on conveyors and fruits sway in the wind. We introduce Static In, Dynamic Out (SIDO), a counterfactual action augmentation that enables a policy trained only on static object demonstrations to adapt to unseen object motion at test time. Our key idea is to factorize moving object manipulation into two sub-problems: predicting where the object will be, and reaching that predicted pose. SIDO displaces the object to a counterfactual future position and morphs the demonstrated action chunk to preserve the hand-object relative pose, yielding a goal-conditioned policy. At deployment an object pose predictor supplies the future position. Across three simulated tasks (Mug, Square, Stack) under five object motion patterns and two real-world tasks (Gantry, Peachtree), SIDO improves moving object success over the baselines while preserving static object performance. Project website: https://sido-staticindynamicout.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。