日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

sim2realarXiv:2006.00057

惑星環境向け移動ロボットのシミュレーションフレームワーク

Simulation Framework for Mobile Robots in Planetary-Like Environments

シェア:XThreadsFacebookLINEはてブBluesky

火星類似環境をGazeboで再現し、ローバーに各種SLAMアルゴリズムを適用して測位性能を評価する枠組みを構築した。

著者: Riccardo Giubilato, Andrea Masili, Sebastiano Chiodini, Marco Pertile, Stefano Debei

分類: cs.RO

原文アブストラクト

In this paper we present a simulation framework for the evaluation of the navigation and localization metrological performances of a robotic platform. The simulator, based on ROS (Robot Operating System) Gazebo, is targeted to a planetary-like research vehicle which allows to test various perception and navigation approaches for specific environment conditions. The possibility of simulating arbitrary sensor setups comprising cameras, LiDARs (Light Detection and Ranging) and IMUs makes Gazebo an excellent resource for rapid prototyping. In this work we evaluate a variety of open-source visual and LiDAR SLAM (Simultaneous Localization and Mapping) algorithms in a simulated Martian environment. Datasets are captured by driving the rover and recording sensors outputs as well as the ground truth for a precise performance evaluation.

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. 3NVIDIA、CoRLでGR00T 2とIsaac Lab 3.0を発表新製品・発表
  4. 4GXO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
  5. 5Google DeepMind、ロボット向けAIを発表新製品・発表

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