Coarse Closed-Loop Trajectory Design of Multiple UAVs for Parallel Data Collection

Multiple unmanned aerial vehicles (UAVs) are able to accelerate the large area data collection procedure by using a parallel manner. Due to the UAV flying ability differentiation and the communication equipment (CE) distribution, the optimal trajectory that minimizes the data collection completion time is difficult to achieve. This paper aims at minimizing the multi-UAV parallel data collection completion time in a wide area without precise CE location information. Different with the existing works, we propose a coarse multi-UAV trajectory design solution without repeated edges, in which each UAV follows a closed-loop trajectory to reduce the data collection time. To facilitate the UAV control and guarantee the data collection of edge CEs, four flight rules are set up. Then a division unit structure with four stay points is proposed to partition the wide area. We prove that a closed-loop trajectory exists for any area that consists of division units with the proposed structure. In the high signal to noise ratio case, the optimal area partition that minimizes the data collection time is obtained. The closed-loop trajectory for each UAV is constructed to minimize the data collection completion time. Simulation results show that the proposed coarse closed-loop trajectory design method is approaching the lower bound of the data collection completion time with a loss less than 10%.

Automated summary

This paper aims to accelerate the large area data collection process by using multiple unmanned aerial vehicles (UAVs) in a parallel manner, and proposes a coarse multi-UAV trajectory design solution without repeated edges to minimize the data collection completion time. The wide area is partitioned using a division unit structure with four stay points, and it is proven that a closed-loop trajectory exists for any area that consists of division units with the proposed structure. Simulation results show that the proposed coarse closed-loop trajectory design method approaches the lower bound of the data collection completion time with a loss less than 10%.