高速画像処理を用いた多指ロボットハンドによる紙面形状のペーパーハンドリング

抄録

Paper handling is an essential part of everyday life and is required in a variety of situations. While paper handling is easy for a human hand, it is difficult for a robot to reproduce the same action every time because the behaviour of paper, which is a flexible material, varies slightly from one paper to another. The aim of this work is to create a motion to manipulate the shape of paper without wrinkling it using high-speed vision, with the aim of replicating human handling of flexible objects with a multi-fingered robotic hand. Two operation patterns were listed, one using both hands to manipulate the paper from both sides and the other using one hand to swing the arm up. The task was successfully completed using image processing with high-speed vision to respond to the paper deforming during the operation.