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 2211.16657cs.ROcs.SYeess.SY
收藏

面向灵巧操作的任务驱动混合模型降阶

Task-Driven Hybrid Model Reduction for Dexterous Manipulation

School for Engineering of Matter, Transport, and Energy at Arizona State University(亚利桑那州立大学材料、运输与能源工程学院);General Robotics, Automation, Sensing and Perception (GRASP) Laboratory, University of Pennsylvania(宾夕法尼亚大学通用机器人、自动化、感知与行动实验室)

Wanxin Jin, Michael Posa

AI总结:

本文提出任务驱动的混合模型降阶方法,从大量接触模式中提取少量任务相关模式,并结合模型预测控制,实现实时高性能的灵巧操作。

摘要:

在接触丰富的任务中,例如灵巧操作,建立和断开接触的混合特性给模型表示与控制带来了挑战。例如,在手中操作中选择接触位置并安排其顺序时,存在数千种潜在的混合模式,通常难以处理。本文受到如下观察的启发:完成许多任务实际所需的模式要少得多。在我们此前学习以线性互补系统表示的混合模型的工作基础上,我们找到了一种仅需有限数量任务相关模式的降阶混合模型。这种简化表示与模型预测控制相结合,可实现实时控制,同时足以取得高性能。我们首先在合成混合系统上验证了所提出的方法,将模式数量降低了多个数量级,同时任务性能损失小于5%。我们还将该方法应用于三指机器人手操作此前未知物体的任务。在没有先验知识的情况下,仅通过收集数千个环境样本,我们在数分钟的在线学习内便实现了最先进的闭环性能。

英文摘要:

In contact-rich tasks, like dexterous manipulation, the hybrid nature of making and breaking contact creates challenges for model representation and control. For example, choosing and sequencing contact locations for in-hand manipulation, where there are thousands of potential hybrid modes, is not generally tractable. In this paper, we are inspired by the observation that far fewer modes are actually necessary to accomplish many tasks. Building on our prior work learning hybrid models, represented as linear complementarity systems, we find a reduced-order hybrid model requiring only a limited number of task-relevant modes. This simplified representation, in combination with model predictive control, enables real-time control yet is sufficient for achieving high performance. We demonstrate the proposed method first on synthetic hybrid systems, reducing the mode count by multiple orders of magnitude while achieving task performance loss of less than 5%. We also apply the proposed method to a three-fingered robotic hand manipulating a previously unknown object. With no prior knowledge, we achieve state-of-the-art closed-loop performance within a few minutes of online learning, by collecting only a few thousand environment samples.

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。

↑