日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

マニピュレーションarXiv:2307.05648

画像処理でロボット把持の堅牢性を高める手法

Image-Processing Based Methods to Improve the Robustness of Robotic Gripping

シェア:XThreadsFacebookLINEはてブBluesky

オプティカルフローを用いた画像処理により、食肉産業などで軟らかく滑りやすい対象物の滑り検出と安定把持を実現するスマートグリッパを提案した論文。

著者: Kristóf Takács, Renáta Nagyné Elek, Tamás Haidegger

分類: cs.RO, cs.SY, eess.SY

原文アブストラクト

Image processing techniques have huge impact on most fields of robotics and industrial automation. Real time methods are usually employed in complex automation tasks, assisting with decision making or directly guiding robots and machinery, while post-processing is usually used for retrospective assessment of systems and processes. While artificial intelligence based image processing algorithms (usually neural networks) are more common nowadays, classical methods can also be used effectively even in modern applications. This paper focuses on optical flow based image processing, proving its efficiency by presenting optical flow based solutions for modern challenges in different fields of robotics such as robotic surgery and food industry automation. The main subject of the paper is a smart robotic gripper designed for automated robot cells in the meat industry, that is capable of slip detection and secure gripping of soft, slippery tissues with the help of the implemented optical flow based algorithm.

関連論文

  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. 5Google DeepMind、ロボット向けAIを発表新製品・発表

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