Path optimization for charging wireless sensor networks using uncrewed aerial vehicles

Full abstract

Read the full abstract

This article investigates an energy supply method for rechargeable wireless sensor networks (WRSNs) utilizing a single uncrewed aerial vehicle (UAV). An UAV-assisted wireless sensor network operating in a one-to-many charging mode is considered, where sensors are powered wirelessly by the UAV. Based on node energy levels, an anchor point selection strategy named Node Energy-Aware Strategy (NEAS) is proposed to minimize the task cycle through energy-weighted clustering. Furthermore, an Enhanced Genetic Algorithm with Ant Colony Optimization-inspired Operators (EGA-ACO) is developed to improve path optimization through pheromone-guided selection, 2-opt local search, and anchor position refinement. Simulation results with 50 to 400 sensor nodes demonstrate that NEAS reduces flight distance by 43.3% and task time by 27.2% compared to the baseline GA algorithm. Building upon NEAS, EGA-ACO achieves an additional 3.0% reduction in flight distance and 11.8% reduction in task time. Collectively, the proposed EGA-ACO achieves 44.9% flight distance reduction and 35.8% task time reduction compared to GA, with advantages becoming more pronounced as network scale increases.