SAIL：模仿学习策略的超演示速度执行
SAIL: Faster-than-Demonstration Execution of Imitation Learning Policies
AI总结：
本文提出SAIL系统，通过一致性保持动作推理、高保真跟踪、自适应速度调制和动作调度，实现模仿学习策略超演示速度执行，在模拟和真实机器人上分别达到4倍和3.2倍加速。
摘要：
离线模仿学习（IL）方法，如行为克隆，在获取复杂的机器人操作技能方面非常有效。然而，现有的IL训练策略仅限于以演示数据中所示的速度执行任务。这限制了机器人系统的任务吞吐量，而这是工业自动化等应用的关键要求。在本文中，我们引入并形式化了实现视觉运动策略超演示速度执行的新问题，并识别了机器人动力学和状态-动作分布转移中的基本挑战。我们将关键见解实例化为SAIL（模仿学习的速度适应），一个集成四个紧密连接组件的全栈系统：（1）用于高速平滑运动的一致性保持动作推理算法，（2）对控制器不变运动目标的高保真跟踪，（3）基于运动复杂度动态调整执行速度的自适应速度调制，以及（4）处理现实世界系统延迟的动作调度。在跨越模拟和两个真实的不同机器人平台的12项任务上的实验表明，SAIL在模拟中实现了高达4倍于演示速度的加速，在现实世界中实现了高达3.2倍的加速。更多细节请访问https://nadunranawaka1.github.io/sail-policy
英文摘要：
Offline Imitation Learning (IL) methods such as Behavior Cloning are effective at acquiring complex robotic manipulation skills. However, existing IL-trained policies are confined to executing the task at the same speed as shown in demonstration data. This limits the task throughput of a robotic system, a critical requirement for applications such as industrial automation. In this paper, we introduce and formalize the novel problem of enabling faster-than-demonstration execution of visuomotor policies and identify fundamental challenges in robot dynamics and state-action distribution shifts. We instantiate the key insights as SAIL (Speed Adaptation for Imitation Learning), a full-stack system integrating four tightly-connected components: (1) a consistency-preserving action inference algorithm for smooth motion at high speed, (2) high-fidelity tracking of controller-invariant motion targets, (3) adaptive speed modulation that dynamically adjusts execution speed based on motion complexity, and (4) action scheduling to handle real-world system latencies. Experiments on 12 tasks across simulation and two real, distinct robot platforms show that SAIL achieves up to a 4x speedup over demonstration speed in simulation and up to 3.2x speedup in the real world. Additional detail is available at https://nadunranawaka1.github.io/sail-policy
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
