Teaching assembly tasks based on semi-autonomous teleoperation for robots with vision and multi-fingered hands

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K12170
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 61050:Intelligent robotics-related
• Research Institution: Wakayama University
• Principal Investigator: Ogawara Koichi 和歌山大学, システム工学部, 准教授 (70452810)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 動作教示 / 遠隔操作 / 組み立て作業

Outline of Final Research Achievements

In order to develop a motion teaching method that can easily teach assembly operations to robots equipped with vision, multi-fingered hands, and robot arms, the following three research tasks were carried out in this study. (1) To realize accurate manipulation of an object by a multi-fingered hand, we developed a technique for manipulating a grasped object by a multi-fingered hand that actively utilizes deformation of elastic elements and a technique for presenting an arbitrary viewpoint image by integrating information from multiple sensors. (2) In order to realize grasping and moving of an object based on three-dimensional information from vision and tactile sensors, we developed techniques for recognizing rigid and non-rigid objects and for motion planning of the robot arm. (3) We developed a simultaneous grasping and object recognition technique to realize learning of teaching actions.

Free Research Field

ロボティクス

Academic Significance and Societal Importance of the Research Achievements

ロボットに搭載されたセンサから得られる情報は限られているため，物体の正確な操作を必要とする作業を遠隔操作で教示することは困難である．本研究では，物体認識に基づく経路探索や物体の安定把持などの機能を自律機能としてロボットに持たせることによって，教示者の負担を大幅に軽減することが可能な技術を開発した．本技術は，たとえば多品種少量生産に対応する多能工ロボットの動作教示などに応用できる．