让机器人在故障时仍能完成任务，靠的是基于扩散模型的自适应轨迹生成。
Moving On, Even When You're Broken: Fail-Active Trajectory Generation via Diffusion Policies Conditioned on Embodiment and Task
- 用扩散模型根据机器人的损伤状态和任务需求生成运动轨迹。
- 无约束任务成功率99.5%，远超基准方法的42.4%。
- 无需重新训练，即可应对训练中未见过的故障情况，适合真实场景部署。
机器人故障往往导致任务中断，需人工干预。本文提出‘故障主动’理念，使机器人在受损时仍能安全完成任务。针对执行器故障，我们提出DEFT——一种基于扩散模型的轨迹生成器，其输入包含机器人当前形态和任务约束。DEFT可泛化至多种故障类型，支持有无约束运动，并在任意故障配置下实现任务完成。我们在7自由度机械臂上进行仿真与真实世界测试，结果表明，在数千种故障条件下，其无约束运动成功率达99.5%，显著优于RRT的42.4%；有约束运动成功率为46.4%，优于微分逆运动学的30.9%。此外，DEFT展现出强大的零样本泛化能力，在训练中未见的故障条件下仍保持高性能。真实世界实验涵盖抽屉操作与擦白板等多步任务，传统方法失败之处，DEFT成功完成。结果证明，DEFT可在各种故障配置与实际部署中实现故障主动操作。
原文摘要 · Abstract (English)
Robot failure is detrimental and disruptive, often requiring human intervention to recover. Our vision is 'fail-active' operation, allowing robots to safely complete their tasks even when damaged. Focusing on 'actuation failures', we introduce DEFT, a diffusion-based trajectory generator conditioned on the robot's current embodiment and task constraints. DEFT generalizes across failure types, supports constrained and unconstrained motions, and enables task completion under arbitrary failure. We evaluate DEFT in both simulation and real-world scenarios using a 7-DoF robotic arm. DEFT outperforms its baselines over thousands of failure conditions, achieving a 99.5% success rate for unconstrained motions versus RRT's 42.4%, and 46.4% for constrained motions versus differential IK's 30.9%. Furthermore, DEFT demonstrates robust zero-shot generalization by maintaining performance on failure conditions unseen during training. Finally, we perform real-world evaluations on two multi-step tasks, drawer manipulation and whiteboard erasing. These experiments demonstrate DEFT succeeding on tasks where classical methods fail. Our results show that DEFT achieves fail-active manipulation across arbitrary failure configurations and real-world deployments.
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。