Real-Time Path-Reconfigurable Coverage Planning for Multi-UAV Missions Over Disjoint Areas
Cai Luo, Jintao Guo, Zixuan Liu, Lei Liu, Chunbo Luo
Abstract
Multi-UAV cooperative coverage missions across geographically separated regions face significant challenges due to potential UAV failures during mission execution. To address the challenges of efficient multi-region coverage and dynamic failure handling, this paper presents a novel real-time path-reconfigurable coverage-planning algorithm for multi-UAV coverage path planning across separated geographical areas with real-time path reconfiguration capabilities. The proposed approach, GRIT-M (Greedy Repair Initializes Tabu search for Multiple regions), extends the GRIT algorithm to efficiently handle both initial planning and online path repair when UAVs fail during mission execution. Unlike existing methods that either focus on single-region coverage or lack failure handling mechanisms, GRIT-M incorporates region-specific knowledge through three key innovations: (1) a composite transition cost function that optimizes inter-region movements, (2) a region-priority greedy repair mechanism that minimizes transition costs during replanning, and (3) a region-preserving Tabu search that maintains area coverage coherence while balancing workload. Extensive simulations demonstrate GRIT-M's scalability and robustness against operational constraints. The algorithm reduces inter-region flights by up to 27% compared to baselines and maintains excellent workload balance (CV <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$< $</tex-math></inline-formula> 7%). Crucially, it shows effective scalability with larger teams (up to 12 UAVs), maintaining a <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$>$</tex-math></inline-formula> 91% replanning success rate with average repair times under 2.1 s. Its resilience to variations in battery capacity, sweep patterns, and initial UAV distributions confirms its suitability for time-critical aerial surveys over disjoint areas.
BibTeX
@inproceedings{ral2026_realtimepathreco,
title = {Real-Time Path-Reconfigurable Coverage Planning for Multi-UAV Missions Over Disjoint Areas},
author = {Cai Luo and Jintao Guo and Zixuan Liu and Lei Liu and Chunbo Luo},
booktitle = {RA-L 2026},
year = {2026}
}