Motor control strategy robust to time-delay and noise of sensory information in sensorimotor integration

2018 Fiscal Year Final Research Report

• Project/Area Number: 16H02882
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Research Field: Intelligent robotics
• Research Institution: Kyushu University
• Principal Investigator: Tahara Kenji 九州大学, 工学研究院, 准教授 (80392033)
• Co-Investigator(Kenkyū-buntansha): 木野 仁 福岡工業大学, 工学部, 教授 (50293816) ; 松谷 祐希 熊本高等専門学校, 機械知能システム工学科, 助教 (80757120) ; 河村 晃宏 九州大学, システム情報科学研究院, 助教 (60706555) ; 越智 裕章 山陽小野田市立山口東京理科大学, 工学部, 助教 (50780128)
• Project Period (FY): 2016-04-01 – 2019-03-31
• Keywords: 時間遅れ / ロバスト / フィードバック / フィードフォワード / 人工筋肉

Outline of Final Research Achievements

A control law of musculoskeletal arm, in which feedback and feedforward are integrated, was proposed and confirmed the effectiveness by simulation and experiment. The robustness was improved by updating the virtual object position in object grasping by a multi-fingered hand, and the usefulness was confirmed by simulation and experiment. Robust sensorless position control was realized for the time delay of the fishing line artificial muscle, and iterative learning control was introduced and confirmed the usefulness by experiments. A force control law by switching and superposition using multiple dielectric elastomer artificial muscles was proposed. Also, a temperature estimation method was proposed, and developed a switching algorithm using the estimated temperature and confirmed the effectiveness by simulation and experiment.

Free Research Field

ロボティクス

Academic Significance and Societal Importance of the Research Achievements

本研究では，センサ情報に含まれる時間遅れやノイズに頑健な制御手法を提案し，安価なセンサやアクチュエータを用いてもある程度の精度が担保される制御を実現した．
人間の運動生成においては，時間遅れやノイズが含まれていたとしても頑健に巧みな制御を実現しており，本研究における成果は，それらを工学的に実現し得る手法の一つとして期待できる．