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:2401.02899

起伏地形におけるマルチUAV探索のためのモデル予測高度・速度制御

Model predictive altitude and velocity control in ergodic potential field directed multi-UAV search

シェア:XThreadsFacebookLINEはてブBluesky

起伏のある地形でのマルチUAV探索において、2次元エルゴディック探索とモデル予測制御を組み合わせ、高度と速度を最適化することで探索効率を高める手法を提案した。

著者: Luka Lanča, Karlo Jakac, Stefan Ivić

分類: cs.RO, cs.MA, math.OC

原文アブストラクト

This research addresses the challenge of executing multi-UAV survey missions over diverse terrains characterized by varying elevations. The approach integrates advanced two-dimensional ergodic search technique with model predictive control of UAV altitude and velocity. Optimization of altitude and velocity is performed along anticipated UAV ground routes, considering multiple objectives and constraints. This yields a flight regimen tailored to the terrain, as well as the motion and sensing characteristics of the UAVs. The proposed UAV motion control strategy is assessed through simulations of realistic search missions and actual terrain models. Results demonstrate the successful integration of model predictive altitude and velocity control with a two-dimensional potential field-guided ergodic search. Adjusting UAV altitudes to near-ideal levels facilitates the utilization of sensing ranges, thereby enhancing the effectiveness of the search. Furthermore, the control algorithm is capable of real-time computation, encouraging its practical application in real-world scenarios.

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

MINISFORUM(ミニPC)

MINISFORUM(ミニPC)の公式広告バナーMINISFORUM公式ストアを見る ↗
PR本紙発行元 EmplifAI