日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

歩行arXiv:2309.16720

粒状地形を歩行する二足歩行ロボットのためのエネルギー効率の良い足形状設計

Energy Efficient Foot-Shape Design for Bipedal Walkers on Granular Terrain

シェア:XThreadsFacebookLINEはてブBluesky

粒状地形上での二足歩行ロボットの運動とエネルギーを解析し、エネルギー効率と歩行効率を考慮した足形状の多目的最適化を行った。硬い粒状地形では非凸形状の足が最良の性能を示すことを発見した。

著者: Xunjie Chen, Jingang Yi, Hao Wang

分類: cs.RO, cs.SY, eess.SY

原文アブストラクト

It is important to understand how bipedal walkers balance and walk effectively on granular materials, such as sand and loose dirt, etc. This paper first presents a computational approach to obtain the motion and energy analysis of bipedal walkers on granular terrains and then discusses an optimization method for the robot foot-shape contour design for energy efficiently walking. We first present the foot-terrain interaction characteristics of the intrusion process using the resistive force theory that provides comprehensive force laws. Using human gait profiles, we compute and compare the ground reaction forces and the external work for walking gaits with various foot shapes on granular terrains. A multi-objective optimization problem is finally formulated for the foot contour design considering energy saving and walking efficiency. It is interesting to find out a non-convex foot shape gives the best performance in term of energy and locomotion efficiency on hard granular terrains. The presented work provides an enabling tool to further understand and design efficient and effective bipedal walkers on granular terrains.

関連論文

  1. 1Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  2. 2GXOとApptronik、Apolloの物流現場試験を拡大 最大1,000台の展開を構想量産・展開
  3. 3newmoとKDDIが資本業務提携 2028年に堺市で自動運転タクシー商用化目指す提携・買収
  4. 4NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  5. 5UC BerkeleyとNVIDIAの研究者、AIエージェントでロボット制御コードを仮想環境で改善研究・技術

PR・アフィリエイト広告

OZ GAMING(GPU搭載BTOゲーミングPC)

OZ GAMING(GPU搭載BTOゲーミングPC)の公式広告バナーOZ GAMINGで完成品PCを見る ↗
PR本紙発行元 EmplifAI