日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

把持/マニピュレーションarXiv:2103.06252

受動的反応を考慮した把持安定性解析

Grasp Stability Analysis with Passive Reactions

シェア:XThreadsFacebookLINEはてブBluesky

クーロン摩擦則の最大散逸原理の非凸性を分解・緩和することで、平面把持の受動安定性を多項式時間で判定する初のアルゴリズムを提案し、空間把持にも階層的緩和で対応した。

著者: Maximilian Haas-Heger

分類: cs.RO

原文アブストラクト

[...] We argue that the traditional grasp modeling theory assumes a complexity that most robotic hands do not possess and is therefore of limited applicability to the robotic hands commonly used today. We discuss these limitations of the existing grasp modeling literature and set out to develop our own tools for the analysis of passive effects in robotic grasping. We show that problems of this kind are difficult to solve due to the non-convexity of the Maximum Dissipation Principle (MDP), which is part of the Coulomb friction law. We show that for planar grasps the MDP can be decomposed into a number of piecewise convex problems, which can be solved for efficiently. [...] Thus, we present the first polynomial runtime algorithm for the determination of passive stability of planar grasps. For the spacial case we [...] describe a convex relaxation of the Coulomb friction law and a successive hierarchical tightening approach that allows us to find solutions to the exact problem including the maximum dissipation principle. [...] The generality of our grasp model allows us to solve a wide variety of problems such as the computation of optimal actuator commands. This makes our framework a valuable tool for practical manipulation applications. Our work is relevant beyond robotic manipulation as it applies to the stability of any assembly of rigid bodies with frictional contacts, unilateral constraints and externally applied wrenches. Finally, we argue that with the advent of data-driven methods as well as the emergence of a new generation of highly sensorized hands there are opportunities for the application of the traditional grasp modeling theory to fields such as robotic in-hand manipulation through model-free reinforcement learning. We present a method that applies traditional grasp models to maintain quasi-static stability throughout a nominally model-free reinforcement learning task. [...]

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2newmoとKDDIが資本業務提携 2028年に堺市で自動運転タクシー商用化目指す提携・買収
  3. 3NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  4. 4GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開
  5. 5UC BerkeleyとNVIDIAの研究者、AIエージェントでロボット制御コードを仮想環境で改善研究・技術

PR・アフィリエイト広告

FRONTIER(GPU搭載BTOパソコン)

FRONTIER(GPU搭載BTOパソコン)の公式広告バナーFRONTIERでGPU搭載PCを見る ↗
PR本紙発行元 EmplifAI