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:2211.03026

3Dビジョンを用いた自動ランデブー・ドッキング

Automated Rendezvous & Docking Using 3D Vision

シェア:XThreadsFacebookLINEはてブBluesky

回転する人工衛星の把持のため、適応カルマンフィルタで3Dビジョンの姿勢推定を高精度化し、遮蔽時でもロボットアームによる把持を可能にした研究。

著者: Farhad Aghili

分類: cs.RO, cs.SY, eess.SY

原文アブストラクト

The robustness and accuracy of a vision system for motion estimation of a tumbling target satellite are enhanced by an adaptive Kalman filter. This allows a vision-guided robot to complete the grasping of the target even if occlusion occurs during the operation. A complete dynamics model, including aspects of orbital mechanics, is incorporated for accurate estimation. Based on the model, an adaptive Kalman filter is developed that estimates not only the system states but also all the model parameters such as the inertia ratio, center-of-mass, and the rotation of the principal axes of the target satellite. An experiment is conducted by using a robotic arm to move a satellite mockup according to orbital mechanics while the satellite pose is measured by a laser camera system. The measurements are sent to the Kalman filter, which, in turn, drives another robotic arm to grasp the target. The results demonstrate successful grasping even if the vision system is blocked for several seconds.

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. 3Google DeepMind、ロボット向けAIを発表新製品・発表
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  5. 5GXO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

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