DexHandDiff：面向自适应灵巧操作的交互感知扩散规划
DexHandDiff: Interaction-aware Diffusion Planning for Adaptive Dexterous Manipulation
浏览论文内容
中文总结 AI 辅助
针对现有扩散规划方法在复杂富接触灵巧操作中存在幽灵状态、适应性不足的问题，提出交互感知框架DexHandDiff，通过双阶段扩散、双重引导及大语言模型辅助生成引导函数，显著提升了训练分布外任务的成功率与泛化性。
中文摘要 AI 辅助
具备丰富接触交互的灵巧操作对高级机器人技术至关重要。尽管近期基于扩散的规划方法在简单操作任务中展现出潜力，但在处理复杂序列交互时，它们往往会产生不切实际的幽灵状态（例如物体无需手部接触就自行移动），或是缺乏适应性。本研究提出DexHandDiff，一种用于自适应灵巧操作的交互感知扩散规划框架。DexHandDiff通过双阶段扩散过程对联合状态-动作动力学建模，该过程包含交互前接触对齐与接触后目标导向控制，可实现目标自适应的泛化性灵巧操作。此外，我们融合了基于动力学模型的双重引导，并利用大语言模型自动生成引导函数，既提升了物理交互的泛化能力，也能通过语言提示实现多样化的目标适配。在开门、笔与方块重定向、物体重定位、锤击等物理交互任务上的实验表明，DexHandDiff在训练分布外的目标上表现出色，平均成功率（59.2% vs 29.5%）是现有方法的两倍以上。该框架在目标自适应灵巧任务上的平均成功率达70.7%，凸显了其在富接触操作中的鲁棒性与灵活性。
英文摘要
Dexterous manipulation with contact-rich interactions is crucial for advanced robotics. While recent diffusion-based planning approaches show promise for simple manipulation tasks, they often produce unrealistic ghost states (e.g., the object automatically moves without hand contact) or lack adaptability when handling complex sequential interactions. In this work, we introduce DexHandDiff, an interaction-aware diffusion planning framework for adaptive dexterous manipulation. DexHandDiff models joint state-action dynamics through a dual-phase diffusion process which consists of pre-interaction contact alignment and post-contact goal-directed control, enabling goal-adaptive generalizable dexterous manipulation. Additionally, we incorporate dynamics model-based dual guidance and leverage large language models for automated guidance function generation, enhancing generalizability for physical interactions and facilitating diverse goal adaptation through language cues. Experiments on physical interaction tasks such as door opening, pen and block re-orientation, object relocation, and hammer striking demonstrate DexHandDiff's effectiveness on goals outside training distributions, achieving over twice the average success rate (59.2% vs. 29.5%) compared to existing methods. Our framework achieves an average of 70.7% success rate on goal adaptive dexterous tasks, highlighting its robustness and flexibility in contact-rich manipulation.
发表机构
- The University of Hong Kong（香港大学）
- University of California, Berkeley（加州大学伯克利分校）
机构由 AI 辅助整理，请以论文原文为准。
