日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

マニピュレーションarXiv:2409.19110

連続体-剛体マニピュレータのためのS-RRT*ベース障害物回避自律動作計画

S-RRT*-based Obstacle Avoidance Autonomous Motion Planner for Continuum-rigid Manipulator

シェア:XThreadsFacebookLINEはてブBluesky

S-RRT*と逆瞬間運動学の零空間を組み合わせ、連続体-剛体マニピュレータが障害物を回避しつつ滑らかな経路を追従する動作計画手法を提案した。

著者: Yulin Li, Tetsuro Miyazaki, Yoshiki Yamamoto, Kenji Kawashima

分類: cs.RO

原文アブストラクト

Continuum robots are compact and flexible, making them suitable for use in the industries and in medical surgeries. Rapidly-exploring random trees (RRT) are a highly efficient path planning method, and its variant, S-RRT, can generate smooth feasible paths for the end-effector. By combining RRT with inverse instantaneous kinematics (IIK), complete motion planning for the continuum arm can be achieved. Due to the high degrees of freedom of continuum arms, the null space in IIK can be utilized for obstacle avoidance. In this work, we propose a novel approach that uses the S-RRT* algorithm to create paths for the continuum-rigid manipulator. By employing IIK and null space techniques, continuous joint configurations are generated that not only track the path but also enable obstacle avoidance. Simulation results demonstrate that our method effectively handles motion planning and obstacle avoidance while generating high-quality end-effector paths in complex environments. Furthermore, compared to similar IIK methods, our approach exhibits superior computation time.

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

MINISFORUM(ミニPC)

MINISFORUM(ミニPC)の公式広告バナーMINISFORUM公式ストアを見る ↗
PR本紙発行元 EmplifAI