Takayuki Murooka

Thank you for your visit. This is my personal page.

Home    About    Projects    CV

Researches

Tool Manipulation with Regrasp Graph

When manipualting tool or grasping tool, there sometimes happens that the tool cannot be moved because of collision or limitation of joint angles of the robot or that the number of the robot's hands are not enough. Then we focus on 'regrasp', which means the motion of changing the grasping pose of the tool. By combining this regrasp with manipulation planning, the robot can manipualte tool more skillfully than ever.

  1. リグラスプ動作を考慮した道具操作計画に基づく自身のネジ締めによる自己修復・拡張
    室岡 貴之，岡田 慧，稲葉 雅幸
    第20回計測自動制御学会システムインテグレーション部門講演会 (SI), 2019 @Kagawa, Japan
    Paper (in Japanese) Poster (in Japanese) Video

  2. Self-Repair and Self-Extension by Tightening Screws based on Precise Calculation of Screw Pose of Self-Body with CAD Data and Graph Search with Regrasping a Driver
    Takayuki Murooka, Kei Okada, and Masayuki Inaba
    The 2019 IEEE-RAS International Conference on Humanoid Robots (Humanoids), 2019 @Toronto, Canada
    Paper Poster Video

  3. 単腕・双腕ロボットにおける平行グリッパによる道具の操作時把持状態へのリグラスプ動作の自動生成
    室岡 貴之，岡田 慧，稲葉 雅幸
    ロボティクス・メカトロニクス講演会2019 (ROBOMECH), 2019 @Hiroshima, Japan
    Paper (in Japanese) Poster (in Japanese)

Tool Manipulation with Large Force

Compatibility of control of position and force is difficult. In particular, when exerting large force is more difficult. By modelling the tool manipulation kinematically, we can distribute force to each limb, which enable tool manipulation with force. Also by using CoG fully, the robot can exert larger force than that with only arms.

  1. Whole-body Posture Generation by Adjusting Tool Force with CoG Movement: Application to Soil Digging
    Takayuki Murooka, Riku Shigematsu, Kunio Kojima, Fumihito Sugai, Yohei Kakiuchi, Kei Okada, and Masayuki Inaba
    The 2019 IEEE-RAS International Conference on Humanoid Robots (Humanoids), 2019 @Toronto, Canada
    Paper Video

  2. 重心を利用した道具作用力の調整による全身動作生成に基づくシャベルを用いた土掘り動作の獲得
    室岡 貴之，重松 陸，小島邦生，菅井 文仁，垣内 洋平，岡田 慧，稲葉 雅幸
    第37回日本ロボット学会学術講演会 (RSJ), 2019 @Tokyo, Japan
    Paper (in Japanese) Video


Tool Manipulation with Unknown Dynamics

There are tasks which cannot be modelled physically because of complex dynamics of the tasks. By approximating the unknown physical model with NN, we can predict the relationship between manipulation to the task and the state of the task. At the same time, by optimizing the input of NN with backpropagation, the robot can aquire optimal manipulation even if the model is redundant to manipulation.

  1. 実時間制御のオンライン学習による双腕台車型ロボットの中国ゴマ操作の獲得
    室岡 貴之，古田 悠貴，岡田 慧，稲葉 雅幸
    第36回日本ロボット学会学術講演会 (RSJ), 2018 @Nagoya, Japan
    Paper (in Japanese) Video

Other Projects

Student Robot Contest

I was a member of RoboTech which is a club aiming to win ABU Robot Contest. I belonged to the department of Control and worked on motion generation including modeling dynamics, path planning, controlling acutuators, parception with some sensors.

  1. NHK Student Robot Contest 2019 @Tokyo, Japan
    • 1st Prize in 24 teams
    I was a chief in the department of Control. After this contest, I retired this club.

  2. ABU Robot Contest 2018 @Ninh Bình, Vietnam
    • quarter-finals Prize in 19 teams
    NHK Student Robot Contest 2018 @Tokyo, Japan
    • 2nd Prize in 24 teams
    I worked on image processing with a Kinect and motion modification with visual data.

  3. ABU Robot Contest 2017 @Bangkok, Thailand
    • 4th Prize in 17 teams
    NHK Student Robot Contest 2017 @Tokyo, Japan
    • 1nd Prize in 24 teams
    I worked on optimization of path for navigation.