日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

週刊ニュースレター購読
arXiv:2103.13245

Anytime informed path re-planning and optimization for robots in changing environments

Anytime informed path re-planning and optimization for robots in changing environments

シェア:XThreadsFacebookLINEはてブBluesky

著者: Cesare Tonola, Marco Faroni, Nicola Pedrocchi, Manuel Beschi

分類: cs.RO

原文アブストラクト

In this paper, we propose a path re-planning algorithm that makes robots able to work in scenarios with moving obstacles. The algorithm switches between a set of pre-computed paths to avoid collisions with moving obstacles. It also improves the current path in an anytime fashion. The use of informed sampling enhances the search speed. Numerical results show the effectiveness of the strategy in different simulation scenarios.