arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2410.00841cs.RO
收藏

面向多指操作任务的扩散信息概率接触搜索

Diffusion-Informed Probabilistic Contact Search for Multi-Finger Manipulation

University of Michigan(密歇根大学);Honda Research Institute USA(美国本田研究所)

Abhinav Kumar, Thomas Power, Fan Yang, Sergio Aguilera Marinovic, Soshi Iba, Rana Soltani Zarrin, Dmitry Berenson

AI总结:

本文提出扩散信息概率接触搜索(DIPS),结合A*搜索与扩散模型,并利用粒子滤波思想处理采样变异性,实现超越训练数据的多指操作接触序列规划。

摘要:

由于动力学的高维性和混合特性,为多指操作规划接触丰富的交互具有挑战性。数据驱动方法的最新进展显示出前景,但对训练数据的质量敏感。将学习与轨迹优化和搜索等经典方法相结合,为问题增加了额外结构,并以约束形式引入领域知识,从而可能超越模型训练所用数据。我们提出扩散信息概率接触搜索(DIPS),该方法利用A*搜索,在扩散模型的信息指导下规划接触模式序列。我们在演示数据集上训练扩散模型,该数据集包含接触模式以及由给定这些模式的轨迹优化器生成的轨迹。此外,我们使用一种受粒子滤波启发的方法来推理由模型误差引起的扩散采样变异性,并利用学习到的判别器估计轨迹的似然。我们表明,我们的方法优于不推理变异性的消融方法,并且能够规划出在多个任务中超越训练数据中所见接触序列的接触序列。我们在模拟桌面卡片滑动和螺丝刀转动任务以及硬件中的螺丝刀任务上进行评估,表明我们结合学习与规划的方法能够迁移到现实世界。

英文摘要:

Planning contact-rich interactions for multi-finger manipulation is challenging due to the high-dimensionality and hybrid nature of dynamics. Recent advances in data-driven methods have shown promise, but are sensitive to the quality of training data. Combining learning with classical methods like trajectory optimization and search adds additional structure to the problem and domain knowledge in the form of constraints, which can lead to outperforming the data on which models are trained. We present Diffusion-Informed Probabilistic Contact Search (DIPS), which uses an A* search to plan a sequence of contact modes informed by a diffusion model. We train the diffusion model on a dataset of demonstrations consisting of contact modes and trajectories generated by a trajectory optimizer given those modes. In addition, we use a particle filter-inspired method to reason about variability in diffusion sampling arising from model error, estimating likelihoods of trajectories using a learned discriminator. We show that our method outperforms ablations that do not reason about variability and can plan contact sequences that outperform those found in training data across multiple tasks. We evaluate on simulated tabletop card sliding and screwdriver turning tasks, as well as the screwdriver task in hardware to show that our combined learning and planning approach transfers to the real world.

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。

↑