池本 周平 (イケモト シュウヘイ)

IKEMOTO Shuhei

写真a

職名

准教授

研究室住所

福岡県北九州市若松区ひびきの2-4

研究分野・キーワード

ロボティクス

ホームページ

http://www.brain.kyutech.ac.jp/~ikemoto/index_ja.html

出身大学 【 表示 / 非表示

  • 2005年03月   金沢大学   工学部   人間機械工学科   卒業   日本国

  • 2003年03月   豊田工業高等専門学校   機械工学科   機械工学科   卒業   日本国

出身大学院 【 表示 / 非表示

  • 2010年03月  大阪大学  工学研究科  知能・機能創成工学専攻  博士課程・博士後期課程  修了  日本国

取得学位 【 表示 / 非表示

  • 大阪大学 -  博士(工学)  2010年03月

学内職務経歴 【 表示 / 非表示

  • 2019年04月
    -
    継続中

    九州工業大学   大学院生命体工学研究科   人間知能システム工学専攻   准教授  

学外略歴 【 表示 / 非表示

  • 2015年04月
    -
    2019年03月

    大阪大学   基礎工学研究科   助教   日本国

  • 2014年07月
    -
    2015年03月

    大阪大学   未来戦略機構第七部門   特任助教   日本国

  • 2010年04月
    -
    2014年06月

    大阪大学   大学院情報科学研究科   助教   日本国

所属学会・委員会 【 表示 / 非表示

  • 2013年08月
    -
    継続中
     

    日本ロボット学会  日本国

 

論文 【 表示 / 非表示

  • Q-bot: heavy object carriage robot for in-house logistics based on universal vacuum gripper

    Isamu Matsuo, Toshihiko Shimizu, Yusuke Nakai, Masahiro Kakimoto, Yuki Sawasaki, Yoshiki Mori, Takamasa Sugano, Shuhei Ikemoto, Takeshi Miyamoto

    Advanced Robotics  ( Taylor & Francis )  34 ( 3-4 ) 173 - 188   2020年01月  [査読有り]

    DOI

  • 2DOF link mechanism mimicking cheetah's spine and leg movement

    Matsumoto O., Shigaki S., Ikemoto S., Chen T.Y., Shimizu M., Hosoda K.

    IEEE International Conference on Robotics and Biomimetics, ROBIO 2019      120 - 125   2019年12月  [査読有り]

     概要を見る

    © 2019 IEEE. A spine of a quadruped animal has compliantly connected several segments. This structure enables the animal to realize bending/stretching motion during running along with the motion of the legs. In this paper, we describe a link mechanism for mimicking such bending/stretching motion of the spine and the leg with few degrees of freedom. By designing a linkage mechanism, we try to realize a similar motion of the spine and the hind leg as that of a cheetah. We developed a prototype with such a structure and demonstrate that it can realize the similar movement of the spine and the leg as that of a cheetah with a simple control strategy.

    DOI Scopus

  • Local Online Motor Babbling: Learning Motor Abundance of a Musculoskeletal Robot Arm

    Liu Z., Hitzmann A., Ikemoto S., Stark S., Peters J., Hosoda K.

    IEEE International Conference on Intelligent Robots and Systems      6594 - 6601   2019年11月  [査読有り]

     概要を見る

    © 2019 IEEE. Motor babbling and goal babbling has been used for sensorimotor learning of highly redundant systems in soft robotics. Recent works in goal babbling have demonstrated successful learning of inverse kinematics (IK) on such systems, and suggest that babbling in the goal space better resolves motor redundancy by learning as few yet efficient sensorimotor mappings as possible. However, for musculoskeletal robot systems, motor redundancy can provide useful information to explain muscle activation patterns, thus the term motor abundance. In this work, we introduce some simple heuristics to empirically define the unknown goal space, and learn the IK of a 10 DoF musculoskeletal robot arm using directed goal babbling. We then further propose local online motor babbling guided by Covariance Matrix Adaptation Evolution Strategy (CMA-ES), which bootstraps on the goal babbling samples for initialization, such that motor abundance can be queried online for any static goal. Our approach leverages the resolving of redundancies and the efficient guided exploration of motor abundance in two stages of learning, allowing both kinematic accuracy and motor variability at the queried goal. The result shows that local online motor babbling guided by CMA-ES can efficiently explore motor abundance at queried goal positions on a musculoskeletal robot system and gives useful insights in terms of muscle stiffness and synergy.

    DOI Scopus

  • Common Dimensional Autoencoder for Learning Redundant Muscle-Posture Mappings of Complex Musculoskeletal Robots

    Masuda H., Hitzmann A., Hosoda K., Ikemoto S.

    IEEE International Conference on Intelligent Robots and Systems      2545 - 2550   2019年11月  [査読有り]

     概要を見る

    © 2019 IEEE. It has been widely considered that a distinctive feature of musculoskeletal structures is that both the joint angle and stiffness can be changed by exploiting the agonistantagonist driving of the joint. However, musculoskeletal systems in animals and humans are typically highly complex, and the simple agonist-antagonist driving is rarely found. Therefore, in accordance with the increasing complexity of musculoskeletal robots, the feature that causes the robot to assume a posture with different stiffness values becomes difficult to achieve, owing to the difficulty in modeling the kinematics. Although datadriven approaches such as the neural network are regarded as suitable for modeling complex relationships, the training data are difficult to obtain because measuring joint stiffness is typically extremely difficult in contrast to measuring an actuator's state and posture. Hence, we herein propose the common dimensional autoencoder where the encoded feature exhibits identical dimensions to the original input vector. In the proposed network, in parallel with the original unsupervised training using the data of the actuators' states, supervised training at part of the encoded features is performed using posture data. Consequently, features expressing the redundancy of inverse kinematics appear at the remaining part of the encoded features without using data such as joint stiffness. The validity of the proposed method was confirmed successfully through an experiment using a 10 degrees-of-freedom complex musculoskeletal robot arm driven by pneumatic artificial muscles.

    DOI Scopus

  • Learning Interactive Behaviors for Musculoskeletal Robots Using Bayesian Interaction Primitives

    Campbell J., Hitzmann A., Stepputtis S., Ikemoto S., Hosoda K., Amor H.B.

    IEEE International Conference on Intelligent Robots and Systems      5071 - 5078   2019年11月  [査読有り]

     概要を見る

    © 2019 IEEE. Musculoskeletal robots that are based on pneumatic actuation have a variety of properties, such as compliance and back-drivability, that render them particularly appealing for human-robot collaboration. However, programming interactive and responsive behaviors for such systems is extremely challenging due to the nonlinearity and uncertainty inherent to their control. In this paper, we propose an approach for learning Bayesian Interaction Primitives for musculoskeletal robots given a limited set of example demonstrations. We show that this approach is capable of real-time state estimation and response generation for interaction with a robot for which no analytical model exists. Human-robot interaction experiments on a 'handshake' task show that the approach generalizes to new positions, interaction partners, and movement velocities.

    DOI Scopus

全件表示 >>