日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

運動計画arXiv:2205.04847

マルチツリー誘導による効率的なロボット運動計画

Multi-Tree Guided Efficient Robot Motion Planning

シェア:XThreadsFacebookLINEはてブBluesky

複数の探索木を同時に育てて探索と活用を両立させ、幾何・微分制約を考慮しつつ高速に軌道を生成する運動計画アルゴリズムを提案した論文。

著者: Zhirui Sun, Jiankun Wang, Max Q. -H. Meng

分類: cs.RO

原文アブストラクト

Motion Planning is necessary for robots to complete different tasks. Rapidly-exploring Random Tree (RRT) and its variants have been widely used in robot motion planning due to their fast search in state space. However, they perform not well in many complex environments since the motion planning needs to simultaneously consider the geometry constraints and differential constraints. In this article, we propose a novel robot motion planning algorithm that utilizes multi-tree to guide the exploration and exploitation. The proposed algorithm maintains more than two trees to search the state space at first. Each tree will explore the local environment. The tree starts from the root will gradually collect information from other trees and grow towards the goal state. This simultaneous exploration and exploitation method can quickly find a feasible trajectory. We compare the proposed algorithm with other popular motion planning algorithms. The experiment results demonstrate that our algorithm achieves the best performance on different evaluation metrics.

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. 2newmo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
  3. 3GXO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
  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

FRONTIER(GPU搭載BTOパソコン)

FRONTIER(GPU搭載BTOパソコン)の公式広告バナーFRONTIERでGPU搭載PCを見る ↗
PR本紙発行元 EmplifAI