A study on temperature-independent material recognition using soft robotic cover embedded in temperature control system

2022 Fiscal Year Final Research Report

• Project/Area Number: 21K21325
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 1002:Human informatics, applied informatics and related fields
• Research Institution: National Institute of Advanced Industrial Science and Technology
• Principal Investigator: Osawa Yukiko 国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 研究員 (80909200)
• Project Period (FY): 2021-08-30 – 2023-03-31
• Keywords: 材料認識 / 熱計測 / 温度制御 / ロボットスキン

Outline of Final Research Achievements

In this research project, a recognition system for contact objects was developed to improve the accuracy of thermal-based material recognition. We investigated approaches that actively utilize heat from both hardware and software perspectives, including developing a temperature-controllable machine (robot) cover and recognition algorithms for contact objects using a time-series temperature database. By adjusting the temperature of the robot gripper according to the temperature of the object, it is possible to prevent deterioration in recognition accuracy due to environmental changes, which was a conventional problem. Furthermore, high-precision classification through machine learning based on temperature information has been achieved for materials with significantly different thermal properties, such as metal and wood, and metals that are difficult to classify using vision and tactile sensors.

Free Research Field

熱工学

Academic Significance and Societal Importance of the Research Achievements

本研究では外装表面の温度制御により、接触物の温度に依存せず材質の特性のみを抽出した熱流の取得を可能とし、それに基づく独自の材質認知手法を構築した。このような温度調節機能が備わった柔軟なロボット外装は未だに実現されておらず、高精度の認識が必要とされる製造・医療現場など，幅広い分野での活用が見込まれる（社会的意義）。また本研究課題を遂行する中で、熱力学・制御工学のみならず情報工学・材料工学など複数の工学分野を統合した新学問領域としての基礎理論の一提案ができたと言える（学術的意義）。