胴体姿勢推定を用いた高速な全身動作生成

ヒューマノイドロボットの全身動作生成に計算時間がかかる問題に対し,事前データから目標の胴体姿勢を推定し,四肢の動作生成計算を軽量化することで,実時間化を狙う.実験では,等身大ヒューマノイドロボットのHRP-4と人間の生活空間を模擬した棚を設置して,合計3秒で屈みながら目標を到達できた.

Publication

  1. Yuya Hakamata, Satoki Tsuichihara, Gustavo Alfonso Garcia Ricardez, Jun Takamatsu, and Tsukasa Ogasawara,
    Pushing and pulling motion generation for humanoid robots using whole body momentum control based on analytical inverse kinematics,
    Advanced Robotics, Vol. 34, No. 21-22, pp. 1442-1454, 2020.
  2. Satoki Tsuichihara, Yuya Hakamata, Gustavo Alfonso Garcia Ricardez, Jun Takamatsu, and Tsukasa Ogasawara,
    Real-time whole-body motion generation using torso posture regression and center of mass,
    ROBOMECH Journal, Vol. 5, No. 1, pp. 1-13, 2018.

深層学習による雑草認識

Satoki Tsuichihara, Shingo Akita, Reiichirou Ike, Masahiro Shigeta, Hiroshi Takemura, Takahiro Natori, Naoyuki Aikawa, Kazumasa Shindo, Yasuyuki Ide, and Shigeki Tejima,
Drone and GPS sensors-based grassland management using deep-learning image segmentation,
In Proceedings of the third IEEE International Conference on Robotic Computing (IRC2019), pp. 608-611, Naples, Italy, Feb 27 2019 (Oral presentation).