Power-Aware Path Planning for Vehicle-Assisted Multi-UAVs in Mobile Crowd Sensing

UAVs’ high mobility and extensive coverage make them widely used as task performers in mobile crowd sensing (MCS). However, due to the limitation of the battery capacity, the flying distance of UAVs is limited, thus they cannot be continuously used in a wide area. In response to this problem, the ground vehicle can be used to transport, release, and recycle UAVs. Large-scale data collection can be achieved through the combined use of the ground vehicle and UAVs, where the route planning of vehicle-assisted UAVs is a key problem. The existing algorithms assume that the power of UAVs is unlimited or the charging time is negligible, which is impractical in real scenarios. In order to solve the above problems, we formalize a vehicle-assisted multi-UAVs path planning model based on the power of UAVs and propose an efficient and power-aware path planning algorithm for vehicle-assisted multi-UAVs(VMUPA). In VMUPA, we take genetic algorithm (GA) to plan flight paths of multi-UAVs under multiple constraints and obtain the power required at each parking spot. Then we optimize the driving route of the ground vehicle according to the remaining power of UAVs to minimize the overall time. Finally, performance evaluation is presented to demonstrate that VMUPA reduces the task completion time by 15% compared to existing algorithms in most cases.