Development of high-precision object shape recognition method using non-contact exploration based on proximity sensors

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K12072
• 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: Kanazawa University
• Principal Investigator: Yosuke Suzuki 金沢大学, フロンティア工学系, 准教授 (20582331)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 近接覚センサ / 点群生成 / ロボットハンド / 形状認識

Outline of Final Research Achievements

In this study, we developed a fingertip sensor and a shape recognition technique for unknown objects to enable a robot to perform dexterous manipulation in place of a human. The fingertip sensor can recognize nearby objects without contact, and can estimate the shape of an object by tracing its circumference while avoiding unintentional contact with the object, as if groping. As a result of our research, we have realized a sensor design that can be used in a variety of applications. We have also developed a method to improve the estimation accuracy by estimating the overall shape in combination with local curvature estimation.

Free Research Field

ロボティクス

Academic Significance and Societal Importance of the Research Achievements

医療・福祉や農業など生活基盤を支える仕事は多くの手作業を含み，ロボットによる代替を行うには未知の物体であっても器用に扱う能力が必要と考えられる．物体を器用に操るには物体の形状を正確に認識する必要がある．本研究成果は，ロボットが指先の感覚を用いて物体の形状を認識することを可能とするものである．指先で探るため視覚のような死角が少なく，また意図せぬ接触を避けられるため安全性にも寄与する．指先のセンサは様々なロボットに実装しやすいモジュール設計を採用しており，汎用性も高いと考えられる．