DemoGrasp：从单次演示实现通用灵巧抓取
DemoGrasp: Universal Dexterous Grasping from a Single Demonstration
AI总结：
DemoGrasp从单次抓取演示出发，将轨迹编辑建模为单步MDP并用RL优化，以简单奖励实现通用灵巧抓取，在仿真和真实世界中均取得领先成功率与强泛化能力。
摘要：
使用多指灵巧手进行通用抓取是机器人操作中的一个基本挑战。尽管最近的方法利用强化学习（RL）成功学习了闭环抓取策略，但高维、长时程探索的固有困难使得复杂的奖励设计和课程设计成为必要，这往往导致在多样化物体上得到次优解。我们提出了DemoGrasp，一种简单而有效的通用灵巧抓取学习方法。我们从抓取特定物体的一条成功演示轨迹出发，通过编辑该轨迹中的机器人动作来适应新物体和新姿态：改变腕部姿态决定在哪里抓取，改变手部关节角度决定如何抓取。我们将这种轨迹编辑形式化为单步马尔可夫决策过程（MDP），并使用RL在仿真中并行优化一个跨数百个物体的通用策略，其奖励仅由一个二值成功项和一个机器人-桌面碰撞惩罚项组成。在仿真中，DemoGrasp使用Shadow Hand在DexGraspNet物体上达到了95%的成功率，优于先前的最先进方法。它还表现出很强的可迁移性，在六个未见过的物体数据集上，跨多种灵巧手本体实现了84.6%的平均成功率，而训练时仅使用了175个物体。通过基于视觉的模仿学习，我们的策略成功抓取了110个未见过的真实世界物体，包括小而薄的物品。它能够泛化到空间、背景和光照变化，同时支持RGB和深度输入，并扩展到杂乱场景中的语言引导抓取。
英文摘要：
Universal grasping with multi-fingered dexterous hands is a fundamental challenge in robotic manipulation. While recent approaches successfully learn closed-loop grasping policies using reinforcement learning (RL), the inherent difficulty of high-dimensional, long-horizon exploration necessitates complex reward and curriculum design, often resulting in suboptimal solutions across diverse objects. We propose DemoGrasp, a simple yet effective method for learning universal dexterous grasping. We start from a single successful demonstration trajectory of grasping a specific object and adapt to novel objects and poses by editing the robot actions in this trajectory: changing the wrist pose determines where to grasp, and changing the hand joint angles determines how to grasp. We formulate this trajectory editing as a single-step Markov Decision Process (MDP) and use RL to optimize a universal policy across hundreds of objects in parallel in simulation, with a simple reward consisting of a binary success term and a robot-table collision penalty. In simulation, DemoGrasp achieves a 95% success rate on DexGraspNet objects using the Shadow Hand, outperforming previous state-of-the-art methods. It also shows strong transferability, achieving an average success rate of 84.6% across diverse dexterous hand embodiments on six unseen object datasets, while being trained on only 175 objects. Through vision-based imitation learning, our policy successfully grasps 110 unseen real-world objects, including small, thin items. It generalizes to spatial, background, and lighting changes, supports both RGB and depth inputs, and extends to language-guided grasping in cluttered scenes.
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。
