摘要
This work examines the path-planning problem for data collection by an unmanned aerial vehicle (UAV) over clusters of smart poles (or streetlights). The smart poles within each cluster are assumed to be interconnected via wired or wireless links, and, thus, the UAV needs to collect data from only one pole in each cluster. A small subset of smart poles is equipped with charging facilities to replenish the UAV's battery when visited. To minimize the UAV's total flight distance in a data collection cycle, we first propose the energy-aware minimum distance (EA-MinDist) path-planning algorithm based on the dynamic programming principle. The algorithm takes into consideration the UAV's need to visit charging poles along the path to sustain its travel over these clusters and can be extended to accommodate an infinite number of data collection cycles while maintaining a constant memory requirement. The sequence of visited smart poles converges to a periodic cycle. Alternatively, to reduce the cost of charging, we also propose the energy-aware minimum charging (EA-MinCharge) algorithm, which aims to minimize the total number of charges needed to traverse these clusters. Simulation results demonstrate the efficacy of the proposed algorithms in comparison to several baseline algorithms.