日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

群制御arXiv:2405.02484

階層分散型異種マルチロボットタスク割り当てシステム

Hierarchically Decentralized Heterogeneous Multi-Robot Task Allocation System

シェア:XThreadsFacebookLINEはてブBluesky

月面などでの資源採掘(ISRU)を想定し、異なる能力を持つ複数ロボットへのタスク割り当てを、階層分散型のオークション方式で調整するシステムを開発し、ROS2上の2Dシミュレータで3つの行動方針を評価した。

著者: Sujeet Kashid, Ashwin D. Kumat

分類: cs.RO

原文アブストラクト

With plans to send humans to the Moon and further, the supply of resources like oxygen, water, fuel, etc., can be satiated by performing In-Situ Resource Utilization (ISRU), where resources from the extra-terrestrial body are extracted to be utilized. These ISRU missions can be carried out by a Multi-Robot System (MRS). In this research, a high-level auction- based Multi-Robot Task Allocation (MRTA) system is developed for coordinating tasks amongst multiple robots with distinct capabilities. A hierarchical decentralized coordination architecture is implemented in this research to allocate the tasks amongst the robots for achieving intentional cooperation in the Multi-Robot System (MRS). 3 different policies are formulated that govern how robots should act in the multiple auction situations of the auction-based task allocation system proposed in this research, and their performance is evaluated in a 2D simulation called pyrobosim using ROS2. The decentralized coordination architecture and the auction-based MRTA make the MRS highly scalable, reliable, flexible, and robust.

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. 2Google DeepMind、ロボット向けAIを発表新製品・発表
  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. 5GXO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

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