U-Auto$^{+}$: Delay-Constrained and Cost-Efficient UAV Air–Ground Network for Remote Inspection
Abstract
Powerline inspection is essential for maintaining the reliability of smart grids by monitoring key components and preventing potential faults. Leveraging uncrewed aerial vehicles (UAVs) for powerline inspection offers superior flexibility and operational efficiency. However, powerlines often span remote and complex terrains, where insufficient network coverage poses significant challenges to long-distance UAV data transmission. In this paper, we investigate the deployment of wireless mesh networks (WMNs) to enable reliable and low-latency communication for long-range UAV inspection tasks. We formulate a novel delay-constrained WMN deployment cost optimization problem considering link characteristics, node heterogeneity, and varied deployment costs. To solve this complex problem, we propose U-Auto<inline-formula><tex-math notation="LaTeX">$^{+}$</tex-math></inline-formula>, a delay-optimized and cost-efficient WMN construction scheme designed to ensure reliable communication for UAV-based inspections. U-Auto<inline-formula><tex-math notation="LaTeX">$^{+}$</tex-math></inline-formula> consists of three key modules: (1) a greedy-based multi-group selection (<italic>GMS</italic>) algorithm for efficient mesh and nest node selection to ensure inspection trajectory coverage, (2) a link probing growth (<italic>LPG</italic>) strategy for constructing globally connected ground WMN topologies, and (3) a proactive routing mechanism supporting two adaptive air-ground link selection schemes: <italic>minDelay</italic>, which minimizes end-to-end latency, and <italic>minSwitch</italic>, which enhances stability by reducing link switching frequency. Based on real geographical data from a forested area in Hubei Province, China, we construct a simulation environment to evaluate U-Auto<inline-formula><tex-math notation="LaTeX">$^{+}$</tex-math></inline-formula>. Extensive experimental results demonstrate that U-Auto<inline-formula><tex-math notation="LaTeX">$^{+}$</tex-math></inline-formula> outperforms baseline methods in network deployment cost, transmission delay, and communication stability.