用人类示范视频训练机器人实时自适应抓取盒子
Latent Adaptive Planner for Dynamic Manipulation
- 在低维隐空间中把规划当作推断，从人示范视频学习策略
- 动态抓盒实验中成功率更高，轨迹更平滑，能耗更低
- 适合需要实时调整动作的机器人操作任务
我们提出潜变量轨迹级策略Latent Adaptive Planner（LAP），用于动态非抓握操作（如盒式抓取），将规划建模为低维隐空间中的推断，并通过人类示范视频有效学习。执行时，LAP通过维护隐计划后验分布并随新观测进行变分重规划，实现实时适应。为弥合人与机器人的具身差距，我们引入基于模型的比例映射方法，从人类示范中重建精确的运动学-动力学关节状态和物体位置。在不同物体属性的挑战性盒式抓取实验中，LAP通过学习类人柔顺运动与自适应行为，展现出更高的成功率、更平滑的轨迹和更高的能量效率。总体而言，LAP实现了动态操作的实时适应，并能使用相同人类示范视频在异构机器人平台间成功迁移。
原文摘要 · Abstract (English)
We present the Latent Adaptive Planner (LAP), a trajectory-level latent-variable policy for dynamic nonprehensile manipulation (e.g., box catching) that formulates planning as inference in a low-dimensional latent space and is learned effectively from human demonstration videos. During execution, LAP achieves real-time adaptation by maintaining a posterior over the latent plan and performing variational replanning as new observations arrive. To bridge the embodiment gap between humans and robots, we introduce a model-based proportional mapping that regenerates accurate kinematic-dynamic joint states and object positions from human demonstrations. Through challenging box catching experiments with varying object properties, LAP demonstrates superior success rates, trajectory smoothness, and energy efficiency by learning human-like compliant motions and adaptive behaviors. Overall, LAP enables dynamic manipulation with real-time adaptation and successfully transfer across heterogeneous robot platforms using the same human demonstration videos.
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。