[论文] Mana: Dexterous Manipulation of Articulated Tools

论文概要

研究领域: CV 作者: Zhao-Heng Yin, Guanya Shi, Pieter Abbeel, C. Karen Liu 发布时间: 2026-06-11 arXiv: 2606.13677

中文摘要

关节工具操作仍然是灵巧机器人领域的一项重大挑战，因为需要协调内部自由度并进行丰富的接触交互。虽然先前的工作主要集中在刚体对象上，但关节工具使用由于其物理复杂性和学习功能性抓取与操作策略的困难仍然未被充分探索。我们提出 Mana（Manipulation Animator），一个通用的 sim-to-real 框架，将灵巧操作重新解释为一个动画问题。受计算机动画启发，Mana 采用粗到细的流水线，通过运动规划和强化学习将过程化生成的抓取关键帧转换为操作轨迹。数据生成过程在很大程度上是自动化的，只需几次鼠标点击即可指定功能可供性（每个工具不到1分钟）。跨越四种不同尺度和关节类型的关节工具，Mana 实现了抓取和手中操作的零样本 sim-to-real 迁移，展示了一种可扩展的灵巧关节工具使用方法。

原文摘要

Articulated tool manipulation remains a major challenge in dexterous robotics due to the need to coordinate internal degrees of freedom and contact-rich interactions. While prior work has largely focused on rigid objects, articulated tool use remains underexplored because of its physical complexity and the difficulty of learning functional grasping and manipulation policies. We present Mana (Manipulation Animator), a general sim-to-real framework that reinterprets dexterous manipulation as an animation problem. Inspired by computer animation, Mana employs a coarse-to-fine pipeline that transforms procedurally-generated grasp keyframes into manipulation trajectories through motion planning and reinforcement learning. The data generation process is largely automatic, requiring only a few mou...

--- *自动采集于 2026-06-15*

#论文 #arXiv #CV #小凯