解决云端执行延迟问题，让智能体在延迟下仍能保持高控制精度。
RAPAC-DP: Response-Aligned Pending-Action Compensation for Diffusion Policies under Delayed Execution
- 用待执行动作序列作为条件，通过轻量补偿路径修正延迟影响。
- 在Kinetix上延迟最大时仍保持81.4%的无延迟性能，RoboMimic任务平均成功率63.3%。
- 无需额外延迟数据即可训练，适合部署在有通信延迟的云端机器人系统。
云端推理使模仿学习策略获得更强计算能力，但通信与计算延迟会降低控制性能。为此，我们提出RAPAC-DP，一种针对扩散模型和流模型动作生成器的响应对齐待执行动作补偿框架。该方法将云响应到达前已安排执行的动作编码为待执行动作序列，并作为条件输入至参数高效的补偿路径。当延迟可忽略时，绕过此路径可精确恢复原始基础策略。训练时，RAPAC-DP从无延迟示范中构建延迟条件样本，无需显式系统动力学或额外延迟示范。在Kinetix上最大固定延迟测试中，保留了81.4%的整体无延迟性能；在每个RoboMimic任务的最大固定延迟下，三个任务平均成功率达到0.633。结果表明，待执行动作补偿对云端部署的模仿学习策略具有显著有效性。
原文摘要 · Abstract (English)
Cloud-side inference gives imitation-learning policies access to greater computational resources, but communication and computation delays can degrade control performance. To compensate for these delays, we propose RAPAC-DP, a response-aligned pending-action compensation framework designed for both diffusion- and flow-based action generators. RAPAC-DP encodes the actions already scheduled for execution before the cloud response arrives into a pending-action sequence that serves as the conditioning input to a parameter-efficient compensation pathway. When delay effects are negligible, bypassing this pathway exactly recovers the frozen base policy. For training, RAPAC-DP constructs delay-conditioned samples from delay-free demonstrations, requiring neither explicit system dynamics nor additional delayed demonstrations. At the largest fixed delay tested on Kinetix, RAPAC-DP retained 81.4% of its overall delay-free performance. At the largest fixed delay tested on each RoboMimic task, it achieved a mean success rate of 0.633 across the three tasks. These results demonstrate the effectiveness of pending-action compensation for cloud-deployed imitation-learning policies.
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。