教机器人像教狗一样：从诱饵、手势和言语中学习敏捷导航
Teaching Robots Like Dogs: Learning Agile Navigation from Luring, Gesture, and Speech
AI总结：
本研究提出了一种人机协同框架，通过手势、言语和诱饵等多模态输入，使机器人高效学习敏捷导航，实验显示在少于1小时的数据下任务成功率高达97.15%。
摘要：
在本工作中，我们旨在使腿部机器人能够通过物理人类指导学习如何解读人类社会线索并产生适当的行为。然而，通过物理互动学习可能会在需要大量人类提供的数据时给用户带来沉重负担。为了解决这个问题，我们提出了一种人机协同框架，使机器人能够以数据高效的方式获取导航行为，并通过多模态自然人类输入进行控制，特别是手势和言语指令。我们使用基于物理的模拟重建交互场景，并汇总数据以缓解由于演示数据有限而产生的分布偏移。我们的渐进式目标提示策略在训练过程中自适应地提供适当的命令和导航目标，从而实现更准确的导航，并增强人类输入与机器人行为之间的对齐。我们在六个现实世界的敏捷导航场景中评估了我们的框架，包括跳过或避开障碍物。我们的实验结果表明，我们的方法在这些场景中几乎所有试验中都成功，总演示数据少于1小时，任务成功率为97.15%。
英文摘要：
In this work, we aim to enable legged robots to learn how to interpret human social cues and produce appropriate behaviors through physical human guidance. However, learning through physical engagement can place a heavy burden on users when the process requires large amounts of human-provided data. To address this, we propose a human-in-the-loop framework that enables robots to acquire navigational behaviors in a data-efficient manner and to be controlled via multimodal natural human inputs, specifically gestural and verbal commands. We reconstruct interaction scenes using a physics-based simulation and aggregate data to mitigate distributional shifts arising from limited demonstration data. Our progressive goal cueing strategy adaptively feeds appropriate commands and navigation goals during training, leading to more accurate navigation and stronger alignment between human input and robot behavior. We evaluate our framework across six real-world agile navigation scenarios, including jumping over or avoiding obstacles. Our experimental results show that our proposed method succeeds in almost all trials across these scenarios, achieving a 97.15% task success rate with less than 1 hour of demonstration data in total.
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。
