Comfortable Control for Active Sheet by Machine Learning

2019 Fiscal Year Final Research Report

• Project/Area Number: 18K13728
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 20020:Robotics and intelligent system-related
• Research Institution: Nagoya University
• Principal Investigator: Shimizu Osamu 名古屋大学, 未来社会創造機構, 特任助教 (90606287)
• Project Period (FY): 2018-04-01 – 2020-03-31
• Keywords: 制御

Outline of Final Research Achievements

I have developed the active sheet by which we can evaluate the comfortability of chairs without prototyping in this research. Hardware of the active sheet is new design without servo motors and their controller. The active sheet can control 32 actuators synchronously. The algorithm of the active sheet controller is based on Simulink, therefore we can revise the algorithm easily. The machine learning algorithm is based on tensorflow, then the controller and machine learning are separated. I evaluates comfortability of sitting by myself due to budget cut for testers.

Free Research Field

制御

Academic Significance and Societal Importance of the Research Achievements

本研究で開発した、座面をアクチュエータにより任意の形状に変化することができるアクティブシートを用いることで、着座時のクッションの柔らかさやクッションが圧縮された後の形状をクッションの試作をすることなく評価が可能になるため、着座の快適性を短期間により精緻に評価できるようになった。今後、評価結果を椅子の設計にフィードバックすることにより、今後より快適性の高い椅子の設計が実現できる可能性がある。