Computing While Navigating: A Novel Task Offloading and Trajectory Planning Scheme for UAV-Assisted MEC System

Unmanned aerial vehicle (UAV)-assisted mobile edge computing (MEC) is a promising solution for delivering high-quality computing services to ground-based Internet of Things devices, where the UAV equipped with an edge server travels among multiple ground devices to provide the required computation services. As computation-intensive tasks become more and more complex, the UAV needs to spend more time at each device for task execution under the traditional working mode, leading to the significant increase of system latency. To well address this issue, in this paper we propose to adopt a novel working mode for the UAV-assisted MEC system, in which the UAV can choose to move to the next device to receive the new task before completing the previous one. Therefore, the UAV can make full use of the travelling time to perform the previous task to improve the system performance. To this end, we present a joint task offloading and UAV trajectory planning optimization problem with the limited battery capacity and tasks latency requirement, aiming to minimize the weighted sum of energy consumption and latency, which is proved to be NP-hard. Then, we propose a hybrid particle swarm optimization based scheme, called ParallUAV, to achieve efficient task offloading and UAV trajectory planning. We conduct extensive simulations to demonstrate the effectiveness of the proposed ParallUAV.