日本フィジカルAI新聞

世界のフィジカルAIを、日本語で。

ドローン/災害対応arXiv:2111.02778

ドローンによる迅速な山火事対応システム

Rapid wildfire response system based on drones

シェア:XThreadsFacebookLINEはてブBluesky

山火事対応のため、移動式緊急オペレーションセンターの配置、ドローン構成の最適化、無線中継ドローンの配置をモデル化し、10年間の適応性を評価した研究。

著者: U IengHou, Bai Yuqian, Luo QiFan

分類: cs.RO

原文アブストラクト

The wildfires in Australia make people suffer the losses caused by the devastating disasters. To protect people's security and the country's development in vulnerable situations, we are expected to establish a set of efficient and sustainable fire-rescuing response mechanisms and tackle the problems faced, including optimization of loadings and quantity of drones, time adaptability of the model, and the locations of radio-repeater drones. We determined the location of mobile emergency operation centers (EOC), which conduct drone launches and deploy rescue personnel. After figuring out where the EOCs are located, we can establish the Drones Planning-programming Budgeting Model to detect the configuration of drones, including optimal numbers and mix of SSA drones and Radio Repeater drones. To estimate the adaptability of our model in the next ten decades, we build a Holt-Winters seasonal model. We utilize the Drones Deployment Optimization model to optimize the locations of hovering VHF/UHF radio-repeater drones. In addition, sensitivity analysis of the model is tested.

  1. 1パナソニックHDら、ロボット操作AI「T-Rex」を共同開発 12タスクで平均成功率65%研究・技術
  2. 2Hugging Face、ロボット学習基盤「LeRobot v0.7.0」を公開 対話操作と大規模データ学習に対応新製品・発表
  3. 3NVIDIA、Vera CPUのエージェント処理性能を公表 x86比で速度・電力効率を比較半導体・GPU
  4. 4Boston Dynamics、車輪付きAtlasの研究試作機を紹介研究・技術
  5. 5AGIBOTのG2 6台、上海のコンビニで24時間試験 1,730タスクを実施量産・展開
PR本紙発行元 EmplifAI