用少于1小时数据教会机器人通过手势语音导航避障
Teaching Robots Like Dogs: Learning Agile Navigation from Luring, Gesture, and Speech
- 通过物理仿真重构交互场景，减少真人示范数据依赖
- 97.15%任务成功率，实现在6种复杂场景下的敏捷避障
- 支持手势+语音多模态控制，适合人机协作场景
本文旨在让双足机器人通过人类物理引导学习理解社交线索并生成恰当行为。传统方法需大量人工示范，负担重。为此，我们提出一种人机协同框架，实现数据高效学习，并支持手势与语音等自然多模态输入。通过物理仿真重建交互场景并聚合数据，缓解示范数据有限导致的分布偏移问题。采用渐进式目标提示策略，在训练中动态提供合适指令与导航目标，提升导航精度与人机行为对齐度。我们在六种真实世界的敏捷导航场景中评估，包括跳跃或避开障碍物。实验结果表明，该方法在所有测试中均表现良好，总示范数据不足1小时，任务成功率达97.15%。
原文摘要 · Abstract (English)
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.
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。