日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

ソフトグリッパ/吸着arXiv:2009.08156

粗く不規則な部品を吸着把持する全軟質ロボットグリッパ

Suction-based Soft Robotic Gripping of Rough and Irregular Parts

シェア:XThreadsFacebookLINEはてブBluesky

平坦な弾性膜を対象物に密着させ、内部の負圧で吸着する全軟質グリッパを提案し、粗い表面でも従来の吸盤の4倍の把持性能を実現した。

著者: Sukho Song, Dirk-Michael Drotlef, Donghoon Son, Anastasia Koivikko, Metin Sitti

分類: physics.app-ph, cs.RO

原文アブストラクト

Recently, suction-based robotic systems with microscopic features or active suction components have been proposed to grip rough and irregular surfaces. However, sophisticated fabrication methods or complex control systems are required for such systems, and robust attachment to rough real-world surfaces still remains a grand challenge. Here, we propose a fully soft robotic gripper, where a flat elastic membrane is used to conform and contact parts or surfaces well, where an internal negative pressure exerted on the air-sealed membrane induces the suction-based gripping. 3D printing in combination with soft molding techniques enable the fabrication of the soft gripper. Robust attachment to complex 3D and rough surfaces is enabled by the surface-conformable soft flat membrane, which generates strong and robust suction at the contact interface. Such robust attachment to rough and irregular surfaces enables manipulation of a broad range of real-world objects, such as an egg, lime, and foiled package, without any physical damage. Compared to the conventional suction cup designs, the proposed suction gripper design shows a four-fold increase in gripping performance on rough surfaces. Furthermore, the structural and material simplicity of the proposed gripper architecture facilitates its system-level integration with other soft robotic peripherals, which can enable broader impact in diverse fields, such as digital manufacturing, robotic manipulation, and medical gripping applications.

  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. 3GXO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
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  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

MINISFORUM(ミニPC)

MINISFORUM(ミニPC)の公式広告バナーMINISFORUM公式ストアを見る ↗
PR本紙発行元 EmplifAI