Verification type general object gripping system based on forced suction gripper of industrial intelligent robot and tactile force sense guided operation

2017 Fiscal Year Final Research Report

• Project/Area Number: 16K12499
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Intelligent robotics
• Research Institution: The University of Tokyo
• Principal Investigator: OKADA Kei 東京大学, 大学院情報理工学系研究科, 教授 (70359652)
• Project Period (FY): 2016-04-01 – 2018-03-31
• Keywords: 知能ロボット / 強制吸引グリッパ / 手探り動作

Outline of Final Research Achievements

Integrated suction hand consisting of forced adsorption and pinching multi fingers was developed, and bending, short range, proximity tactile sensor was incorporated here. It shows that the product can be taken out from the narrow space by adding the extendable shaft to the suction part, and by using multi-fingered pinching, the article is prevented from falling. In addition, we designed an algorithm to design primitive reflecting motion according to the foreshock response obtained via the near range tactile sensor, thereby realizing trial - and - error grasping along the shape of the object. In addition, by acquiring a recognition behavior network to predict failure before learning, applying only the hand attached operation of the double arm only when the failure probability is high, thereby suppressing the extension of the work execution time , It became possible to secure high operational stability.

Free Research Field

知能ロボティクス